DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
The amendments filed on 09/22/2025 has been entered. The status of the claims is as follow:
Claims 1-20 remain pending in the application.
Claims 1-20 are rejected.
Response to Arguments
In reference to the rejections under 35 USC 101:
Argument:
Applicant asserts in Remarks pg. 11-12 that the amendments made following the telephonic interviews of November 5, 2024 and December 17, 2024 add structural features sufficient to overcome the rejection under 35 U.S.C 101. Specifically, Applicant relies on newly added features reciting conversion of signals and adjustment of quantum states of qubits of a quantum device as allegedly integrating the claim subject matter into a practical applications.
Response:
Examiner respectfully disagrees and notes that the relied-upon claim amendments introduce new matter that is not supported by the originally filed specification, in violation of 35 U.S.C 112(a). As previously set forth, the original disclosure does not describe or suggest that the conversion of signals is performed for the purpose of adjusting quantum states of qubits, nor does it provide sufficient structural, functional or operational detail explaining how the converted signals interact with qubits to manipulate or adjust their quantum states within a quantum device. The Specification, as originally files, discusses optimization problems and signal processing at a high level, but fails to disclose any concrete implementation linking such signal conversion to the physical control or adjustment of quantum states of qubits. The newly added limitations therefore go beyond mere clarification or elaboration and instead introduce new functional relationships and technical effects that are not reasonably conveyed to a person of ordinary skill in the art from the original disclosure. Because the amendments introduce subject matter that lacks written description support, the amended limitations are not entered for purpose of overcoming the 101 rejection. Consequently, Applicant’s arguments relying on these newly added features are not persuasive.
Applicant’s arguments filed on 09/22/2025 have been fully considered but they are not persuasive.
In reference to the rejections under 35 USC 112(a) – New Subject Matter:
Argument 1: Applicant asserts in Remarks pg. 12-13 that the specification reasonably conveys possession of the newly added feature “thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced”, on the basis that classical computing devices inherently operate using electrical signals. Applicant further argues that, because the first set of data is introduced by one or more classical computing device, the problem is necessarily defined by electrical signals.
Examiner respectfully disagrees and notes that while it is generally understood that classical computing devices operate using electrical signals, the written description requirement under 35 U.S.C 112(a) is not satisfied by general knowledge or inherent properties alone. Rather, the specification must reasonably convey to one of ordinary skill in the art that the inventor had possession of the specific claimed subject matter as of the filling date (see Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1340, 94 USPQ2d 1161, 1167 (Fed. Cir. 2010)). In the present application, the originally filed specification does not describe or suggest that the electrical signals obtained from the classical computing device defining a problem using a first set of data. The reference to signals is made solely in the context of adapting them for transmission to the quantum device, without indicating any semantic or representational relationship between the signals and a defined dataset. The claimed limitation introduces specific functional content related to problem definition and data representation in electrical signals, which is not supported by the instant specification. Applicant’s reliance on the inherent operation of classical computing devices improperly attempts to supply missing disclosure through general technical knowledge. However, new matter cannot be supported by reasoning that the claimed feature would have been obvious or implicit based on how devices typically function, absent an explicit or implicit disclosure in the specification typing that functionality to the claimed invention. Moreover, the newly added limitation goes beyond merely acknowledging that classical computers use electrical signals, it affirmatively recites a specific act of “obtaining electrical signals” and characterizes those signals as defining the problem. This is a substantive technical detail that was not originally disclosed in the specification as filed.
Argument 2: Applicant asserts in Remarks pg. 13-14 that the originally filed specification supports the newly added limitation “converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device”, relying on the disclosure of converting electrical signals into light or another type of signals adapted to be received by the quantum device.
Examiner respectfully disagrees and notes that while the specification discloses that electrical signals from one or more classical computing devices may be converted into light or another type of signals to enable communication or interfacing with the quantum device, the specification does not describe or suggest that the purpose of such signal conversion is to adjust or control quantum states of qubits. The cited paragraph from Specification describes signal conversion in the context of communications compatibility between classical computing devices and the quantum device, not as a mechanism for qubit state manipulation. Specifically, the disclosure indicates that the converter adapts electrical signals into a form that the quantum device is adapted to receive. However, the specification fails to describe how the converted signals functionally operate to adjust quantum states of qubits, or that such adjustment is the intended result of the signal conversion. There is no description of control signals or any technical details that would reasonably convey possession of using the converted signals to adjust quantum states. Applicant’s assertion that quantum devices solve problems by adjusting quantum states of qubits reflects general knowledge in the art, but this general understanding cannot supply missing written description for a specific claimed functional relationship between the converted signals and qubit state adjustment. The written description requirement is not satisfied by inferring undisclosed functionality based on how quantum devices generally operate (see Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1340, 94 USPQ2d 1161, 1167 (Fed. Cir. 2010))
Argument 3:Applicant asserts in Remarks pg. 14-15 that the originally filed specification supports the limitation of “adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced”, on the basis that quantum devices generally operate by interaction between signals and qubits to adjust quantum states.
Examiner respectfully disagrees and notes that although quantum devices generally operate through interactions between signals and qubits, the specification as originally filed does not disclose sufficient details describing adjusting quantum states of qubits within the quantum device, nor does it describe that such adjustment is performed by interaction between “another type of signals” and the qubits as now claimed. The specification lacks any description of the nature of the signals, the manner of their interaction with the qubits, or how such interaction is used to adjust quantum states for solving the problem. Applicant’s reliance on general principles of quantum device operation improperly supplies missing disclosure through general knowledge rather than through the specification itself. Accordingly, the originally filed specification does not reasonably convey possession of the claimed feature at the time of filling.
Argument 4: Applicant asserts in Remarks pg. 15 that because the problem is solved using a quantum device, and quantum devices inherently operate by adjusting quantum states of qubits, the claimed feature “the quantum device with the adjusted quantum states of qubits” is necessarily supported by the originally filed specification.
Examiner respectfully disagrees and notes that although quantum devices generally operate by adjusting quantum states of qubits, the specification as originally filed does not disclose or describe “the quantum device with the adjusted quantum states of qubits” as a claimed feature of the invention. The specification does not explain which quantum states are adjusted, how the states are adjusted or in what manner the adjusted quantum states are associated with solving the problem. Applicant’s assertion relies on the general principle that quantum devices function by adjusting qubit states. However, general knowledge regarding how quantum devices operate cannot substitute for an express or implicit disclosure in the specification demonstrating possession of the claimed subject matter. The written description requirement requires that the specification itself reasonably convey to one of ordinary skill in the art that the inventor had possession of the claimed feature at the time of filling.
Applicant’s arguments filed 09/22/2025 have been fully considered but they are not persuasive.
In reference to the rejections under 35 USC 103:
Applicant asserts in Remarks pg. 15-23 that Cao fails to discloses using an optimized portfolio for training a machine learning algorithm.
Examiner respectfully disagrees and notes that as shown in Cao, the reference explicitly formulates a portfolio optimization problem that is solved using a deep learning model trained with a defined loss function. In particular, Cao discloses that the main goals is to “estimate the optimal solution wopt for the portfolio optimization problem” in order to “obtain the maximum Sharpe ratio during the next K days” and further states that “to study a deep learning model for the portfolio optimization problem, we aim at only using the historical stock data during the las M days for training and then predict the optimal equally weighted portfolio” (see discussion following Eq(3)). Moreover, Section 2.2 of Cao explicitly introduces “a new loss function for the Sharpe-ratio maximization” and explains that this loss function is implemented within “an appropriate deep neural network to estimate the optimal solution w” (Equation (5)). The output of this trained neural network is then used to derive the final optimized portfolio vector wopt according to the thresholding rule in Eq. (6). Thus, Cao clearly teaches that the portfolio optimization objective (i.e., Sharpe-ratio-based optimized portfolio weights) is embedded directly into the training of the deep learning model via the loss function. Contrary to Applicant’s assertion, Cao does not merely compute an optimized portfolio as a post-processing step independent of machine learning. Rather, the optimization of the portfolio is integrated into the training process of the deep neural network itself, where historical stock data are used as training inputs and the loss function explicitly encodes the portfolio optimization objective. As such, Cao discloses using an optimized portfolio formulation for training a machine learning algorithm, namely by training a deep neural network to output portfolio selection vectors that maximize the Sharpe ratio under defined constraints.
Applicant’s arguments filed on 09/22/2025 have been fully considered but they are not persuasive.
Claim Rejections - 35 USC § 112(a) – New Matter
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
The newly amended limitation of independent claims 1, 11 and 20 “thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced”, introduces new subject matter that is not supported by the instant specification. The closest discussion in the specification discloses communication between classical computing devices and quantum device, and describes converting electrical signals into other forms of signals (e.g., light) to enable communication (see pg. 8, lines 7-14 of the Instant Specification). However, the paragraph does not describe or suggest that the electrical signals obtained from the classical computing device defining a problem using a first set of data. The reference to signals is made solely in the context of adapting them for transmission to the quantum device, without indicating any semantic or representational relationship between the signals and a defined dataset. The claimed limitation introduces specific functional content related to problem definition and data representation in the electrical signals, which is not supported by the general disclosure of signal conversion for communication purpose. The Examiner acknowledges that the Applicant has cited some of the paragraphs in the specification as the support for the newly amended limitations. However, the Examiner has fully considered all the cited paragraphs and found that those cited paragraphs in the specification do not support the newly amended elements of the independent claims 1, 11 and 20. Applicant is required to cancel the new matter in the reply to this Office Action.
The newly amended limitation of independent claims 1, 11 and 20 “converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device”, introduces new subject matter that is not supported by the instant specification. The closest discussion in the specification discloses that a converter may adapt electrical signals of a computing device into light or other types of signals for communication with a quantum device (e.g., when the quantum device is photonic) (see pg. 8, lines 7-14 of the Instant Specification). However, it does not describe or suggest that the purpose of the signal conversion is to adjust quantum states of qubits. The specification lacks sufficient detail to support the functional use of the converted signals for manipulating or adjusting the quantum states of qubits within the quantum device. The specification at no time discusses the “quantum states of qubits of a quantum device”, let alone adjusting the quantum states of qubits within the quantum device. The Examiner acknowledges that the Applicant has cited some of the paragraphs in the specification as the support for the newly amended limitations. However, the Examiner has fully considered all the cited paragraphs and found that those cited paragraphs in the specification do not support the newly amended elements of the independent claims 1, 11 and 20. Applicant is required to cancel the new matter in the reply to this Office Action.
The newly amended limitation of independent claims 1, 11 and 20 “adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced”, introduces new subject matter that is not supported by the instant specification. The closest discussion in the specification discloses obtaining the optimal trading trajectory upon solving the cost function (see pg. 3, lines 20-21 of the Instant Specification). However, it does not describe or suggest adjusting quantum states of the qubits for solving the problem with the first set of data introduced. The specification lacks sufficient detail to support the manipulating or adjusting the quantum states of qubits within the quantum device. The specification at no time discusses the “quantum states of qubits of a quantum device”, let alone adjusting the quantum states of qubits for solving the problem with the first set of data introduced. The Examiner acknowledges that the Applicant has cited some of the paragraphs in the specification as the support for the newly amended limitations. However, the Examiner has fully considered all the cited paragraphs and found that those cited paragraphs in the specification do not support the newly amended elements of the independent claims 1, 11 and 20. Applicant is required to cancel the new matter in the reply to this Office Action.
The newly amended limitation of independent claims 1, 11 and 20 “the quantum device with the adjusted quantum states of qubits”, introduces new subject matter that is not supported by the instant specification. The closest discussion in the specification refers to a variational quantum circuit comprising several qubits coupled via entangling gates and including one qubit gates with variational parameters to be optimized using a conjugate gradient algorithm (see pg. 10, lines 16-18 of the Instant Specification). However, this disclosure is limited to describing the structural components and the presence of tunable parameters in the quantum circuit. It does not explicitly or implicitly describe an operation or process that adjusts the quantum states of the qubits, nor does it explain how the quantum states are modified or the purpose of such adjustment. The specification lacks sufficient detail to support the manipulating or adjusting the quantum states of qubits within the quantum device. The specification at no time discusses the “quantum states of qubits of a quantum device”, let alone adjusting the quantum states of qubits. The Examiner acknowledges that the Applicant has cited some of the paragraphs in the specification as the support for the newly amended limitations. However, the Examiner has fully considered all the cited paragraphs and found that those cited paragraphs in the specification do not support the newly amended elements of the independent claims 1, 11 and 20. Applicant is required to cancel the new matter in the reply to this Office Action.
Dependent claims 2-10, and 9-19 inherit the deficiencies of the independent claims 1, 11. Therefore they are also rejected under the same rationale.
The newly amended limitation of dependent claim 9 “wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits, the resulting quantum states defining values of binary variables of the quadratic unconstrained binary optimization problem, the values of the binary variables defining the optimal trading trajectories for the first period of time”, introduces new subject matter that is not supported by the instant specification. The closest discussion in the specification discloses obtaining the optimal trading trajectory upon solving the cost function, wherein the cost function includes a plurality of binary variables (see pg. 3, lines 20-21 of the Instant Specification). However, it does not describe or suggest performing further adjustments of the quantum states of the qubits to obtain the resulting quantum states defining values of binary variables of the quadratic unconstrained binary optimization problem. The specification lacks sufficient detail to support the manipulating or adjusting the quantum states of qubits within the quantum device to determine the resulting quantum states of the qubits. The specification at no time discusses the “quantum states of qubits of a quantum device”, let alone adjusting the quantum states of qubits to obtain the resulting quantum states defining values of binary variables of the quadratic unconstrained binary optimization problem. The Examiner acknowledges that the Applicant has cited some of the paragraphs in the specification as the support for the newly amended limitations. However, the Examiner has fully considered all the cited paragraphs and found that those cited paragraphs in the specification do not support the newly amended elements of dependent claim 9. Applicant is required to cancel the new matter in the reply to this Office Action.
The newly amended limitation of dependent claim 10 “wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits; the method comprising converting the resulting quantum states of the qubits into electrical signals; wherein inputting the optimal trading trajectories comprises receiving, by the one or more classical computing devices, the electrical signals obtained by converting the resulting quantum states of the qubits; and wherein the training of the machine learning algorithm is performed by the one or more classical computing devices”, introduces new subject matter that is not supported by the instant specification. The closest discussion in the specification discloses obtaining the optimal trading trajectory upon solving the cost function, wherein the cost function includes a plurality of binary variables (see pg. 3, lines 20-21 of the Instant Specification), and digitally training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the second period of time (see pg. 3, lines 25-30 of the Instant Specification). However, it does not describe or suggest performing further adjustments of the quantum states of the qubits to obtain the resulting quantum states defining values of binary variables of the quadratic unconstrained binary optimization problem, nor does it disclose the electrical signals are obtained by converting the resulting quantum states of the qubits. The specification lacks sufficient detail to support the manipulating or adjusting the quantum states of qubits within the quantum device to determine the resulting quantum states of the qubits. The specification at no time discusses the “quantum states of qubits of a quantum device”, let alone adjusting the quantum states of qubits to obtain the resulting quantum states defining values of binary variables of the quadratic unconstrained binary optimization problem. Additionally, the Examiner further submits the specification does not disclose that the inputting of the optimal trading trajectories comprises receiving the electrical signals obtained by converting the resulting quantum states of the qubits. The specification at no time discusses the “resulting quantum states of”, let alone converting the resulting quantum states of the qubits.The Examiner acknowledges that the Applicant has cited some of the paragraphs in the specification as the support for the newly amended limitations. However, the Examiner has fully considered all the cited paragraphs and found that those cited paragraphs in the specification do not support the newly amended elements of the dependent claim 10. Applicant is required to cancel the new matter in the reply to this Office Action.
Claim Rejections - 35 USC § 101 - Abstract Ideas
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1 – 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding claim 1,
Step 1 – Is the claim directed to a process, machine, manufacture, or composition of matter? – Yes, the claim is directed to a process.
Step 2A – Prong 1 – Does the claim recite an abstract idea, law of nature, or natural phenomenon? – Yes, the claim recites the abstract ideas:
providing a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio – This limitation is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time– This limitation is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
provides a recommended composition of an asset portfolio based on a set of inputs - This limitation is directed to the abstract idea of a mental process (including an observation, evaluation, judgement, opinion) which can be performed in the human mind, or by a human using pen and paper (see MPEP 2106.04(a)(2) Ill. C.)
Step 2A – Prong 2 – Does the claim recite any additional elements that integrate the judicial exception into a practical application? – No, there are no additional elements that integrate the judicial exception into a practical application.
digitally providing – This limitation is directed to a recitation of the words “apply it” (or an equivalent) with the judicial exception, such as mere instructions to implement an abstract idea on a computer [See MPEP 2106.05(f)] and therefore fails to integrate the judicial exception into a practical application.
digitally introducing – This limitation is directed to a recitation of the words “apply it” (or an equivalent) with the judicial exception, such as mere instructions to implement an abstract idea on a computer [See MPEP 2106.05(f)] and therefore fails to integrate the judicial exception into a practical application.
Introducing, [by one or more classical computing devices], a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets – This limitation is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
by one or more classical computing devices This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced This limitation is directed to insignificant extra-solution
activity - mere data gathering (see MPEP 2106.05(g)).
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device; - Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced; Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
digitally providing a quantum or classical machine learning algorithm This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
digitally training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the first period of time and minimizing a predetermined error function for each time unit of a first period of time for which there is historical financial data in the first set of data – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
introducing a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
providing a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
Step 2B – Does the claim recite additional elements that amount to significantly more than the judicial exception? – No, there are no additional elements that amount to significantly more than the judicial exception.
digitally providing – This limitation is directed to a recitation of the words “apply it” (or an equivalent) with the judicial exception, such as mere instructions to implement an abstract idea on a computer [See MPEP 2106.05(f)] and therefore fails to amount to significantly more than the judicial exception.
digitally introducing – This limitation is directed to a recitation of the words “apply it” (or an equivalent) with the judicial exception, such as mere instructions to implement an abstract idea on a computer [See MPEP 2106.05(f)] and therefore fails to amount to significantly more than the judicial exception.
Introducing, [by one or more classical computing devices], a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets – This limitation is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
by one or more classical computing devices This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced This limitation is directed to insignificant extra-solution
activity - mere data gathering (see MPEP 2106.05(g)).
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device; - Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced; Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
digitally providing a quantum or classical machine learning algorithm This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2))
digitally training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the first period of time and minimizing a predetermined error function for each time unit of a first period of time for which there is historical financial data in the first set of data – This limitation is directed to training a machine learning algorithm by using an input data set and minimizing an error function, which is well-understood, routine, and conventional activity as explained by Camacho et al., (US PGPUB No US2007/0042718 A1) (“In FIG. 6 the training model of a generic neural network (601) is described. To be able to carry out the adjustment of the internal parameters, weightings and offsets of the various neurons which constitute the neural network (500, 601), it is necessary to have an input data set (602) and the targets (603)
which the network has to attain for said inputs. There are well known algorithms which permit the network to be trained minimizing the error (605) between the output values (604) and the targets (603) which a comparator (606) provides.” [0091]) and therefore fails to amount to significantly more than the judicial exception.
introducing a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to amount to significantly more than the judicial exception.
providing a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to amount to significantly more than the judicial exception.
Regarding claim 2,
Step 2A – Prong 1 – The claim recites the additional abstract ideas:
solving the quadratic unconstrained binary optimization problem for the second period of time [with the quantum device], thereby obtaining optimal trading trajectories for the second period of time – This limitation is further limiting the solving the quadratic unconstrained binary optimization problem limitation from claim 1, and is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
Step 2A – Prong 2 – The claim recites the additional elements:
further comprising, after the historical financial data is available for the second period of time: digitally introducing a third set of data into the problem, the third set of data comprising historical financial data for the second period of time – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
solving the quadratic unconstrained binary optimization problem for the second period of time with the quantum device – This limitation is further limiting the quantum device limitation from claim 1, and is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
digitally training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the second period of time minimizing a predetermined error function for each time unit of the second period of time for which there is historical financial data in the third set of data – This limitation is further limiting the training the machine learning algorithm limitation from claim 1, and is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
digitally introducing a fourth set of data into the machine learning algorithm, the fourth set of data comprising financial data for a third period of time that is posterior to the second period of time – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
and digitally providing a recommended portfolio composition for the third period of time by running the trained machine learning algorithm with the fourth set of data introduced therein – This limitation is further limiting the trained machine learning algorithm limitation of claim 1, and is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
Step 2B – The claim recites the additional elements:
further comprising, after the historical financial data is available for the second period of time: digitally introducing a third set of data into the problem, the third set of data comprising historical financial data for the second period of time – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.].
solving the quadratic unconstrained binary optimization problem for the second period of time with the quantum device – This limitation is further limiting the quantum device limitation from claim 1, and is directed to solving a quadratic unconstrained binary optimization problem with a quantum device, which is well-understood, routine, and conventional activity as explained by Glover et al., (Quantum Bridge Analytics I: A Tutorial on Formulating and Using QUBO Models) (“we focus on the Quadratic Unconstrained Binary Optimization (QUBO) model which is presently the most widely applied optimization model in the quantum computing area” [Page 1, Abstract]).
digitally training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the second period of time minimizing a predetermined error function for each time unit of the second period of time for which there is historical financial data in the third set of data – This limitation is further limiting the machine learning algorithm limitation of claim 1, and is directed to training a machine learning algorithm by using an input data set and minimizing an error function, which is well-understood, routine, and conventional activity as explained by Camacho et al., (US PGPUB No US2007/0042718 A1) (“In FIG. 6 the training model of a generic neural network (601) is described. To be able to carry out the adjustment of the internal parameters, weightings and offsets of the various neurons which constitute the neural network (500, 601), it is necessary to have an input data set (602) and the targets (603) which the network has to attain for said inputs. There are well known algorithms which permit the network to be trained minimizing the error (605) between the output values (604) and the targets (603) which a comparator (606) provides.” [0091]).
digitally introducing a fourth set of data into the machine learning algorithm, the fourth set of data comprising financial data for a third period of time that is posterior to the second period of time – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.].
and digitally providing a recommended portfolio composition for the third period of time by running the trained machine learning algorithm with the fourth set of data introduced therein – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.].
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 3,
Step 2A – Prong 1 – The claim recites the additional abstract idea:
further comprising digitally commanding making one or more investments based on the recommended portfolio composition provided – This limitation is directed to the abstract idea of a mental process (concepts performed in the human mind, including observation and evaluation [see MPEP 2106.04(a)(2) III. C.]). The digitally commanding is reciting a computer at a high level of generality which merely uses a computer as a tool to perform the concept.
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 4,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the machine learning algorithm comprises a neural network or a variational quantum circuit – This limitation is further limiting the machine learning algorithm limitation of claim 1, and is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
Step 2B – The claim recites the additional element:
wherein the machine learning algorithm comprises a neural network or a variational quantum circuit – This limitation is further limiting the machine learning algorithm limitation of claim 1, and is directed to implementing a machine learning algorithm as a neural network, which is well-understood, routine, and conventional activity as explained by Cao et al. (DELAFO: An Efficient Portfolio Optimization Using Deep Neural Networks), “deep neural networks such as e.g. convolutional neural networks (CNN) and recurrent neural networks (RNN) have been proven to work well in many applications and multi-variable time series data” [Page 624, 1 Introduction]).
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 5,
Step 2A – Prong 1 – The claim recites the additional abstract idea:
wherein the cost function is
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A at least comprises the following terms
PNG
media_image2.png
55
383
media_image2.png
Greyscale
or at least comprises the following terms
PNG
media_image3.png
47
534
media_image3.png
Greyscale
, where wt is a vector of the components of which are the percentages of each asset in the portfolio at time t, μt is a vector of expected returns at time t, γ is a parameter controlling the volatility of the portfolio, ∑t is a matrix of covariances of the returns at time t, v-t is a percentage of transaction costs, ∆wt is a change in the composition of the vector of assets between time t and time t + 1, Λt is a matrix of market impact at time t, and ti and tf are an initial time and a final time of a respective period of time – This limitation is further limiting the cost function limitation of claim 1, and is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 6,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the quantum device comprises one of: a quantum annealer, a hybrid quantum-classical machine, a universal gate-based quantum computer, or a Gaussian Boson Sampling quantum device – This limitation is further limiting the quantum device limitation of claim 1, and is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
Step 2B – The claim recites the additional element:
wherein the quantum device comprises one of: a quantum annealer, a hybrid quantum-classical machine, a universal gate-based quantum computer, or a Gaussian Boson Sampling quantum device – This limitation is further limiting the quantum device limitation of claim 1, and is directed to implementing an algorithm onto a quantum device, which is well-understood, routine, and conventional activity as explained by Kerman (US PGPUB No US2017/0141286 A1), “The two most well-known quantum-processing paradigms are: digital quantum computing … and quantum annealing” [0003].
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 7,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the digital steps are carried out with the one or more classical computing devices – This limitation invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)].
Step 2B – The claim recites the additional element:
wherein the digital steps are carried out with one or more computing devices – This limitation invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)].
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 8,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the one or more classical computing devices comprise one or more of: a computer processing unit, a graphics processing unit, and a field-programmable gate array – This limitation is further limiting the computing devices limitation from claim 7, and invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)].
Step 2B – The claim recites the additional element:
wherein the one or more classical computing devices comprise one or more of: a computer processing unit, a graphics processing unit, and a field-programmable gate array – This limitation is further limiting the computing devices limitation from claim 7, and invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)].
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 9,
Step 2A – Prong 2 – The claim recites the additional element:
wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits, This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
the resulting quantum states defining values of binary variables of the quadratic unconstrained binary optimization problem, This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
the values of the binary variables defining the optimal trading trajectories for the first period of time This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
Step 2B – The claim recites the additional element:
wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits, This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
the resulting quantum states defining values of binary variables of the quadratic unconstrained binary optimization problem, This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
the values of the binary variables defining the optimal trading trajectories for the first period of time This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 10,
Step 2A – Prong 2 – The claim recites the additional element:
wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits; This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
the method comprising converting the resulting quantum states of the qubits into electrical signals; Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
wherein inputting the optimal trading trajectories comprises receiving, by the one or more classical computing devices, the electrical signals obtained by converting the resulting quantum states of the qubits; and This limitation is directed to insignificant extra-solution activity (see MPEP 2106.05(g)).
wherein the training of the machine learning algorithm is performed by the one or more classical computing devices Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
Step 2B – The claim recites the additional element:
wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits; This claim merely recites a further limitation on the solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time from Claim 1 which was directed to mathematical calculation (see MPEP 2106.04(a)(2) l. C.)
the method comprising converting the resulting quantum states of the qubits into electrical signals; Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
wherein inputting the optimal trading trajectories comprises receiving, by the one or more classical computing devices, the electrical signals obtained by converting the resulting quantum states of the qubits; and This limitation is directed to receiving or transmitting data over a network. The courts have recognized receiving or transmitting data over a network as well understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity (see MPEP 2106.05(d) II.).
wherein the training of the machine learning algorithm is performed by the one or more classical computing devices Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 11,
Step 1 – Is the claim directed to a process, machine, manufacture, or composition of matter? – Yes, the claim is directed to an apparatus.
Step 2A – Prong 1 – Does the claim recite an abstract idea, law of nature, or natural phenomenon? – Yes, the claim recites the abstract ideas:
providing a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio – This limitation is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
solve the quadratic unconstrained binary optimization problem for the first period of time, thereby obtaining optimal trading trajectories for the first period of time;– This limitation is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
provides a recommended composition of an asset portfolio based on a set of inputs - This limitation is directed to the abstract idea of a mental process (including an observation, evaluation, judgement, opinion) which can be performed in the human mind, or by a human using pen and paper (see MPEP 2106.04(a)(2) Ill. C.)
Step 2A – Prong 2 – Does the claim recite any additional elements that integrate the judicial exception into a practical application? – No, there are no additional elements that integrate the judicial exception into a practical application.
comprising: a quantum device; and one or more computing devices communicatively coupled with the quantum device; the one or more computing devices being configured to at least cause the apparatus to – This limitation invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)] and therefore fails to integrate the judicial exception into a practical application.
introduce a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced This limitation is directed to insignificant extra-solution activity - mere data gathering (see MPEP 2106.05(g)).
the signal converter being configured to This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
convert the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of the quantum device Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
the quantum device being configured to This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
adjust, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
the quantum device, with the adjusted quantum states of the qubits, being configured to at least cause the apparatus to solve the quadratic unconstrained binary optimization problem for the first period of time – This limitation is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
the one or more computing devices being configured to at least further cause the apparatus to– This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
train the machine learning algorithm by both inputting the optimal trading trajectories obtained by both inputting the optimal trading trajectories obtained by the quantum device for the first period of time and minimizing a predetermined error function for each time unit of the first period of time for which there is historical data in the first set of data – This limitation is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
introduce a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
provide a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein – This limitation is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
Step 2B – Does the claim recite additional elements that amount to significantly more than the judicial exception? – No, there are no additional elements that amount to significantly more than the judicial exception.
comprising: a quantum device; and one or more computing devices communicatively coupled with the quantum device; the one or more computing devices being configured to at least cause the apparatus to – This limitation invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)] and therefore fails to amount to significantly more than the judicial exception.
introduce a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to amount to significantly more than the judicial exception.
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced This limitation is directed to insignificant extra-solution activity - mere data gathering (see MPEP 2106.05(g)).
the signal converter being configured to This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
convert the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of the quantum device Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
the quantum device being configured to This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
adjust, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
the quantum device, with the adjusted quantum states of the qubits, being configured to at least cause the apparatus to solve the quadratic unconstrained binary optimization problem for the first period of time – This limitation is directed to solving a quadratic unconstrained binary optimization problem with a quantum device, which is well-understood, routine, and conventional activity as explained by Glover et al., (Quantum Bridge Analytics I: A Tutorial on Formulating and Using QUBO Models) (“we focus on the Quadratic Unconstrained Binary Optimization (QUBO) model which is presently the most widely applied optimization model in the quantum computing area” [Page 1, Abstract]) and therefore fails to amount to significantly more than the judicial exception.
the one or more computing devices being configured to This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
train the machine learning algorithm by both inputting the optimal trading trajectories obtained by both inputting the optimal trading trajectories obtained by the quantum device for the first period of time and minimizing a predetermined error function for each time unit of the first period of time for which there is historical financial data in the first set of data – This limitation is directed training a machine learning algorithm by using an input data set and minimizing an error function, which is well-understood, routine, and conventional activity as explained by Camacho et al., (US PGPUB No US2007/0042718 A1) (“In FIG. 6 the training model of a generic neural network (601) is described. To be able to carry out the adjustment of the internal parameters, weightings and offsets of the various neurons which constitute the neural network (500, 601), it is necessary to have an input data set (602) and the targets (603) which the network has to attain for said inputs. There are well known algorithms which permit the network to be trained minimizing the error (605) between the output values (604) and the targets (603) which a comparator (606) provides.” [0091]) and therefore fails to amount to significantly more than the judicial exception.
introduce a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to amount to significantly more than the judicial exception.
provide a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to amount to significantly more than the judicial exception.
Regarding claim 12,
Step 2A – Prong 1 – The claim recites the additional abstract ideas:
after historical financial data is available for the second period of time, solve the quadratic unconstrained binary optimization problem for the second period of time, thereby obtaining optimal trading trajectories for the second period of time – This limitation is further limiting the solve the quadratic unconstrained binary optimization problem from claim 11, and is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
Step 2A – Prong 2 – The claim recites the additional elements:
wherein: the one or more computing devices are configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, introduce a third set of data into the problem, the third set of data comprising historical financial data for the second period of time – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
the quantum device is configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, solve the quadratic unconstrained binary optimization problem for the second period of time – This limitation is further limiting the quantum device limitation from claim 11, and is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
the one or more computing devices are configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, train the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the second period of time and minimizing a predetermined error function for each time unit of the second period of time for which there is historical financial data in the third set of data – This limitation is further limiting the train the machine learning algorithm limitation from claim 11, and is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
introduce a fourth set of data into the machine learning algorithm, the fourth set of data comprising financial data for a third period of time that is posterior to the second period of time – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
provide a recommended portfolio composition for the third period of time by running the trained machine learning algorithm with the fourth set of data introduced therein – This limitation is further limiting the trained machine learning algorithm limitation of claim 11, and is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
Step 2B – The claim recites the additional elements:
wherein: the one or more computing devices are configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, introduce a third set of data into the problem, the third set of data comprising historical financial data for the second period of time – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.].
the quantum device is configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, solve the quadratic unconstrained binary optimization problem for the second period of time – This limitation is further limiting the quantum device limitation from claim 11, and is directed to solving a quadratic unconstrained binary optimization problem with a quantum device, which is well-understood, routine, and conventional activity as explained by Glover et al., (Quantum Bridge Analytics I: A Tutorial on Formulating and Using QUBO Models) (“we focus on the Quadratic Unconstrained Binary Optimization (QUBO) model which is presently the most widely applied optimization model in the quantum computing area” [Page 1, Abstract]).
the one or more computing devices are configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, train the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the second period of time and minimizing a predetermined error function for each time unit of the second period of time for which there is historical financial data in the third set of data – This limitation is further limiting the train the machine learning algorithm limitation from claim 11, and is directed to training a machine learning algorithm by using an input data set and minimizing an error function, which is well-understood, routine, and conventional activity as explained by Camacho et al., (US PGPUB No US2007/0042718 A1) (“In FIG. 6 the training model of a generic neural network (601) is described. To be able to carry out the adjustment of the internal parameters, weightings and offsets of the various neurons which constitute the neural network (500, 601), it is necessary to have an input data set (602) and the targets (603) which the network has to attain for said inputs. There are well known algorithms which permit the network to be trained minimizing the error (605) between the output values (604) and the targets (603) which a comparator (606) provides.” [0091]).
introduce a fourth set of data into the machine learning algorithm, the fourth set of data comprising financial data for a third period of time that is posterior to the second period of time – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.].
provide a recommended portfolio composition for the third period of time by running the trained machine learning algorithm with the fourth set of data introduced therein – This limitation is further limiting the trained machine learning algorithm limitation of claim 11, and is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.].
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 13,
Step 2A – Prong 1 – The claim recites the additional abstract idea:
wherein the one or more computing devices are configured to at least further cause the apparatus to command making one or more investments based on the recommended portfolio composition provided – This limitation is directed to the abstract idea of a mental process (concepts performed in the human mind, including observation and evaluation [see MPEP 2106.04(a)(2) III. C.]). The one or more computing devices are recited at a high level of generality which merely uses a computer as a tool to perform the concept.
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 14,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the machine learning algorithm comprises a neural network or a variational quantum circuit – This limitation is further limiting the machine learning algorithm limitation of claim 11, and is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
Step 2B – The claim recites the additional element:
wherein the machine learning algorithm comprises a neural network or a variational quantum circuit – This limitation is further limiting the machine learning algorithm limitation of claim 11, and is directed to implementing a machine learning algorithm as a neural network, which is well-understood, routine, and conventional activity as explained by Cao et al. (DELAFO: An Efficient Portfolio Optimization Using Deep Neural Networks), “deep neural networks such as e.g. convolutional neural networks (CNN) and recurrent neural networks (RNN) have been proven to work well in many applications and multi-variable time series data” [Page 624, 1 Introduction]).
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 15,
Step 2A – Prong 1 – The claim recites the additional abstract idea:
wherein the cost function is
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A at least comprises the following terms
PNG
media_image2.png
55
383
media_image2.png
Greyscale
or at least comprises the following terms
PNG
media_image3.png
47
534
media_image3.png
Greyscale
, where wt is a vector of the components of which are the percentages of each asset in the portfolio at time t, μt is a vector of expected returns at time t, γ is a parameter controlling the volatility of the portfolio, ∑t is a matrix of covariances of the returns at time t, v-t is a percentage of transaction costs, ∆wt is a change in the composition of the vector of assets between time t and time t + 1, Λt is a matrix of market impact at time t, and ti and tf are an initial time and a final time of a respective period of time – This limitation is further limiting the cost function limitation of claim 11, and is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 16,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the quantum device comprises one of: a quantum annealer, a hybrid quantum-classical machine, a universal gate-based quantum computer, or a Gaussian Boson Sampling quantum device – This limitation is further limiting the quantum device limitation of claim 1, and is directed to is directed to insignificant extra-solution activity [see MPEP 2106.05(g)].
Step 2B – The claim recites the additional element:
wherein the quantum device comprises one of: a quantum annealer, a hybrid quantum-classical machine, a universal gate-based quantum computer, or a Gaussian Boson Sampling quantum device – This limitation is further limiting the quantum device limitation of claim 1, and is directed to implementing an algorithm onto a quantum device, which is well-understood, routine, and conventional activity as explained by Kerman (US PGPUB No US2017/0141286 A1), “The two most well-known quantum-processing paradigms are: digital quantum computing … and quantum annealing” [0003].
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 17,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the one or more classical computing devices comprise one or more of: a computer processing unit, a graphics processing unit, and a field-programmable gate array – This limitation is further limiting the computing devices limitation from claim 7, and invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)].
Step 2B – The claim recites the additional element:
wherein the one or more computing devices comprise one or more of: a computer processing unit, a graphics processing unit, and a field-programmable gate array – This limitation is further limiting the computing devices limitation from claim 11, and invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)].
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 18,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the first period of time comprises a plurality of days and the second period of time comprises one day – This limitation is further limiting the first period of time and second period of time limitations of claim 11, and is directed to insignificant extra-solution activity (selecting a particular data source or type of data to be manipulated [see MPEP 2106.05(g)]).
Step 2B – The claim recites the additional element:
wherein the first period of time comprises a plurality of days and the second period of time comprises one day – This limitation is further limiting the first period of time and second period of time limitations of claim 11, and is directed to insignificant extra-solution activity (selecting a particular data source or type of data to be manipulated [see MPEP 2106.05(g)]).
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 19,
Step 2A – Prong 2 – The claim recites the additional element:
wherein the one day of the second period of time is today or yesterday – This limitation is further limiting the second period of time comprises one day limitation of claim 9, and is directed to insignificant extra-solution activity (selecting a particular data source or type of data to be manipulated [see MPEP 2106.05(g)]).
Step 2B – The claim recites the additional element:
wherein the one day of the second period of time is today or yesterday – This limitation is further limiting the second period of time comprises one day limitation of claim 9, and is directed to insignificant extra-solution activity (selecting a particular data source or type of data to be manipulated [see MPEP 2106.05(g)]).
Thus, the judicial exception is not integrated into a practical application [see MPEP 2106.05(d) I.], failing Step 2A Prong 2. The claim does not include any additional elements that are sufficient to amount to significantly more than the judicial exception, under step 2B.
Regarding claim 20,
Step 1 – Is the claim directed to a process, machine, manufacture, or composition of matter? – Yes, the claim is directed to a process.
Step 2A – Prong 1 – Does the claim recite an abstract idea, law of nature, or natural phenomenon? – Yes, the claim recites the abstract ideas:
providing a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio – This limitation is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
solving the quadratic unconstrained binary optimization problem for the first period of time with the[[a]] quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time– This limitation is directed to a mathematical concept (mathematical formulas or equations [see MPEP 2106.04(a)(2) I. B.]).
provides a recommended composition of an asset portfolio based on a set of inputs - This limitation is directed to the abstract idea of a mental process (including an observation, evaluation, judgement, opinion) which can be performed in the human mind, or by a human using pen and paper (see MPEP 2106.04(a)(2) Ill. C.)
Step 2A – Prong 2 – Does the claim recite any additional elements that integrate the judicial exception into a practical application? – No, there are no additional elements that integrate the judicial exception into a practical application.
A non-transitory computer-readable medium encoded with instructions that, when executed by at least one processor or hardware, make an apparatus at least perform the following – This limitation invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)] and therefore fails to integrate the judicial exception into a practical application.
introducing ,[by one or more classical computing devices], a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
by one or more classical computing devices This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced This limitation is directed to insignificant extra-solution activity - mere data gathering (see MPEP 2106.05(g)).
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device; - Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced; Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
providing a quantum or classical machine learning algorithm This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the first period of time and minimizing a predetermined error function for each time unit of a first period of time for which there is historical financial data in the first set of data – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
introducing a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
providing a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein – This limitation is directed to insignificant extra-solution activity [see MPEP 2106.05(g)] and therefore fails to integrate the judicial exception into a practical application.
Step 2B – Does the claim recite additional elements that amount to significantly more than the judicial exception? – No, there are no additional elements that amount to significantly more than the judicial exception.
A non-transitory computer-readable medium encoded with instructions that, when executed by at least one processor or hardware, make an apparatus at least perform the following – This limitation invokes a computer merely as a tool for performing an existing process [see MPEP 2106.05(f)(2)] and therefore fails to amount to significantly more than the judicial exception.
Introducing, by one or more classical computing devices, a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to integrate the judicial exception into a practical application.
by one or more classical computing devices This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced This limitation is directed to insignificant extra-solution activity - mere data gathering (see MPEP 2106.05(g)).
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device; - Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced; Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea [see MPEP 2106.05(f)] and therefore fails to integrate the exception into a practical application.
providing a quantum or classical machine learning algorithm This limitation is directed to a computer merely used as a tool to perform an existing process (see MPEP 2106.05(f) (2)).
training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the first period of time and minimizing a predetermined error function for each time unit of a first period of time for which there is historical financial data in the first set of data – This limitation is directed to *, which is well-understood, routine, and conventional activity as explained by ** and therefore fails to amount to significantly more than the judicial exception.
introducing a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to integrate the judicial exception into a practical application.
providing a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein – This limitation is directed to receiving or transmitting data over a network, which the courts have recognized as well-understood, routine, conventional activity when they are claimed at a high level of generality or as insignificant extra-solution activity [see MPEP 2106.05(d) II. i.] and therefore fails to integrate the judicial exception into a practical application.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1 – 2, 4, 6 – 12, 14, 16 – 18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Cao et al. (DELAFO: An Efficient Portfolio Optimization Using Deep Neural Networks) (hereafter referred to as Cao) in view of Johnson et al. (US PGPUB No US 2017/0372427 A1) (hereafter referred to as Johnson) and further in view of Choi (US 8,190,548 B2)
Regarding claim 1, Cao discloses a method comprising:
digitally introducing a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets (Cao: “To study a deep learning model for the portfolio optimization problem, we aim at only using the historical stock data during the last M days for training … we represent each input data as a 3D tensor … that includes all stock data (both the volume and the price) of N different tickers during the last M consecutive days” [*Examiner note: the training data is being interpreted as a first set of data, and the last M days are being interpreted as a first period of time] [Page 626, 2.1 Problem Formation]);
digitally providing a quantum or classical machine learning algorithm that provides a recommended composition of an asset portfolio based on a set of inputs (Cao: “To estimate the output vector w, we consider different deep learning approaches for solving the portfolio optimization… we construct a new ResNet architecture for the problem and create four other combinations of deep neural networks. They are SA + LSTM (Self-Attention model and LSTM), SA + GRU (Self-Attention and GRU)” [*Examiner note: the Self-Attention and GRU deep neural network algorithm is being interpreted as a provided classical machine learning algorithm that solves the portfolio optimization (i.e., provides a recommended composition of an asset portfolio)] [Page 627 – 628, 2.3 Our Proposed Models for the Portfolio Optimization]);
digitally training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the [quantum] device for the first period of time and minimizing a predetermined error function for each time unit of the first period of time for which there is historical financial data in the first set of data (Cao: “one can estimate the maximum value of the Sharpe ratio by solving the following optimization problem … one can use the stochastic gradient descent method for approximating the optimal w” [*Examiner note: i.e., w is being interpreted as the optimal trading trajectories] [Page 627, 2.2 A New Loss Function for the Sharpe-Ratio Maximization] … “we apply ResNet for estimating the optimal value for the vector w in the loss function” [*Examiner note: i.e., estimating the optimal value for a loss function is being interpreted as minimizing a predetermined error function for w (i.e., input optimal trading trajectories)] [Page 628, 2.3 Our Proposed Models for the Portfolio Optimization]);
digitally introducing a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets (Cao: “On each day, we collect the information of both “price” and “volume” of these 381 tickers and y, the daily return on the market in the next K days” [*Examiner note: i.e., the daily return on the market for the next K days is being interpreted as a second set of data for a second period of time] [Page 631, 3.2 Data Preparation]);
and digitally providing a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein (Cao: “The main goal in our work is to estimate the optimal solution wopt for the portfolio optimization problem (3) in order to obtain the maximum Sharpe ratio during the next K days … we aim at only using the historical stock data during the last M days as training and then predict the optimal equally weighted portfolio having the highest Sharpe ratio during the next K days.” [Page 626, 2.1 Problem Formation]).
Cao fails to disclose:
That the device is a quantum device
digitally providing a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio;
by one or more classical computing devices
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced;
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device;
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced;
solving the quadratic unconstrained binary optimization problem for the first period of time with a quantum device with the adjusted quantum states of the qubits, thereby obtaining optimal trading trajectories for the first period of time;
However, Johnson discloses:
That the device is a quantum device (Johnson: “FIG. 1 is a diagram illustrating a financial portfolio optimization method implemented using a quantum processing device” [0006])
by one or more classical computing devices (Johnson, ¶[0033]: “The platform library 145 is not necessarily limited to interacting with just quantum processing devices. One or more classical solver libraries 160 for conventional processing devices may also be used by the platform library 145 for various purposes (e.g. to solve some part of a problem that is not well-suited to a quantum processing solution, or to compare an answer obtained on a quantum processing device to an answer obtainable via a classical solver library).”)
digitally providing a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio (Johnson: “An objective function is formulated based on the list of lots and on the budget. The objective function has a form suitable for solution by a quantum processing device, which is used to optimize the objective function” [Abstract] … “The third term is based on a component Gcost for the cost or budget allocated to purchase assets.” [*Examiner note: i.e., the objective function contains a cost function Gcost] [0020] … “Module 220 uses heuristics to convert a higher-order polynomial binary optimization problem into a quadratic binary optimization problem [*Examiner note: i.e., the objective function is a quadratic unconstrained binary optimization problem] [0047]);
solving the quadratic unconstrained binary optimization problem for the first period of time with a quantum device, thereby obtaining optimal trading trajectories for the first period of time (Johnson: “Given the financial offerings 10, the method runs on a digital computer 20 and queries the market to download historical market data 30 to computer the daily rate of return, its variance, and its covariance over a period of time specified by the portfolio manager” [*Examiner note: emphasis added] [0014] … “Once the portfolio manager chooses 40 his targeted level of expected rate of return for the entire portfolio, the objective function 50 can be reconstructed to optimize for a desired expected rate of return and minimum variance. The problem is resubmitted to the quantum annealing computer 60 to determine the selection of lots to purchase that meet the budget, expected rate of return, and minimum variance in price” [0016] … “The result 70 is output to the portfolio manager showing a spectrum of solutions including the optimum along with a number of near-optimum solutions for consideration.” [*Examiner note: i.e., returning the optimum solution is solving the objective (i.e., quadratic unconstrained binary optimization) function to obtain optimal trading trajectories] [0017]);
Cao and Johnson both disclose inventions relating to the field of endeavor relating to portfolio optimization and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to replace the optimal trajectories approximated by Cao with the optimal trajectories generated by the quadratic unconstrained binary optimization problem taught by Johnson. One having ordinary skill in the art would have been motivated to make this change before the effective filing date of the claimed invention because quantum devices are faster than traditional computer for portfolio optimization (Johnson: “This method formulates the portfolio optimization problem and solves for the best solutions using the power of quantum annealing computers to achieve solutions much faster than existing methods running on traditional digital computer hardware alone” [0012]).
However, Choi explicitly discloses:
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced; (Choi, Col. 14, Lines 40-43: “In some cases, analog processor interface module 530 may communicate with driver module 532 rather than directly with NIC 540 in order to send and receive signals from quantum processor 550.”)
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device; (Choi, Col. 14, Lines 26-36: “Where computing system 500 includes a driver module 532, the driver module 532 may include instructions to output signals to quantum processor 550. NIC 540 may include appropriate hardware required for interfacing with qubit nodes 552 and coupling devices 554, either directly or through readout device 556, qubit control system 558, and/or coupling device control system 560. Alternatively, NIC 540 may include software and/or hardware that translate commands from driver module 532 into signals (e.g., voltages, currents, optical signals, etc.) that are directly applied to qubit nodes 552 and coupling devices 554.”)
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced; (Choi, Col. 5, Lines 31-47: “The respective bias may control a tunneling rate of each of the physical qubits. The respective bias may control a height of a potential barrier between a first state and a second state of the physical qubit. Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state. Increasing the height of the potential barrier may ensure a state of the physical qubit does not change from the first state to the second state. At least two variables from the Quadratic Unconstrained Binary Optimization problem may be assigned to two respective logical 40 qubit. Each of the respective logical qubits may include at least two of the physical qubits coupled by at least one of the physical qubit couplers. Each pair from the at least two variables may have a relationship in the Quadratic Unconstrained Binary Optimization problem which is represented by a 45 respective controllable coupling between a respective physical qubit from each of the respective logical qubits.”)
with the adjusted quantum states of the qubits (Choi, Col. 5, Lines 34-36: “Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state.”)
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Cao and Choi. Cao teaches a method for the optimal trajectories approximated. Choi teaches methods for analog processing using quantum computing devices to solve Quadratic Unconstrained Binary Optimization problems. One of ordinary skill would have motivation to combine Cao and Choi to enable direct physical interaction with the quantum hardware, allowing qubit states to be tuned according to the encoded problem parameters.
Regarding claim 2, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
further comprising, after historical financial data is available for the second period of time: digitally introducing a third set of data into the problem, the third set of data comprising historical financial data for the second period of time (Cao: “we use the time windows of M consecutive days for extracting the input data of proposed models. On each day, we collect the information of both “price” and “volume“ of these 381 tickers and y, the daily return on the market in the next K days (K = 19). Consequently, the input data has the shape (381, 64, 2) and we move the time window during the studying period of time (from January 1, 2013, to July 31, 2019) to obtain 1415 samples.” [*Examiner note: moving the time window from the first M days forward to the next M days would encompass the next K days from the previous iteration, i.e., this third data set would be the historical data for the next K days (i.e., the second period of time)] [Page 631, 3.2 Data Preparation]);
solving the quadratic unconstrained binary optimization problem for the second period of time with the quantum device, thereby obtaining optimal trading trajectories for the second period of time (Johnson: “Given the financial offerings 10, the method runs on a digital computer 20 and queries the market to download historical financial data 30 to compute the daily rate of return, its variance, and its covariance over a period of time specified by the portfolio manager.” [*Examiner note: i.e., the manager can specify the second period of time] [0014] … “The problem is resubmitted to the quantum annealing computer 60 to determine the selection of lots to purchase that meet the budget, expected rate of return, and minimum variance in price” [0016] … “The result 70 is output to the portfolio manager showing a spectrum of solutions including the optimum along with a number of near-optimum solutions for consideration.” [*Examiner note: i.e., returning the optimum solution is solving the objective (i.e., quadratic unconstrained binary optimization) function to obtain optimal trading trajectories] [0017]);
digitally training the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the second period of time and minimizing a predetermined error function for each time unit of the second period of time for which there is historical financial data in the third set of data (Cao: “only using the historical stock data during the last M days for training or updating the proposed model, and then doing prediction for the equally weighted portfolio having the highest Sharpe ratio during the next K days” [*Examiner note: emphasis added. i.e., after the algorithm has been trained on data from the first window, it is updated on data from the following windows] [Page 625, 1 Introduction]);
digitally introducing a fourth set of data into the machine learning algorithm, the fourth set of data comprising financial data for a third period of time that is posterior to the second period of time (Cao: “The main goal in our work is to estimate the optimal solution wopt for the portfolio optimization problem (3) in order to obtain the maximum Sharpe ratio during the next K days … we aim at only using the historical stock data during the last M days as training and then predict the optimal equally weighted portfolio having the highest Sharpe ratio during the next K days.” [Page 626, 2.1 Problem Formation] … we move the time window during the studying period of time (from January 1, 2013, to July 31, 2019) to obtain 1415 samples.” [*Examiner note: these windows include a fourth set of data that comprises financial data for a third period of time posterior to the second period of time] [Page 631, 3.2 Data Preparation]);
and digitally providing a recommended portfolio composition for the third period of time by running the trained machine learning algorithm with the fourth set of data introduced therein (Cao: “The main goal in our work is to estimate the optimal solution wopt for the portfolio optimization problem (3) in order to obtain the maximum Sharpe ratio during the next K days … we aim at only using the historical stock data during the last M days as training and then predict the optimal equally weighted portfolio having the highest Sharpe ratio during the next K days.” [Page 626, 2.1 Problem Formation]).
Regarding claim 4, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein the machine learning algorithm comprises a neural network or a variational quantum circuit (Cao: “by implementing an appropriate deep neural network to estimate the optimal solution …, we can derive the final output vector wopt” [Page 627, 2.2 A New Loss Function for the Sharpe-Ratio Maximization]).
Regarding claim 6, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein the quantum device comprises one of: a quantum annealer, a hybrid quantum-classical machine, a universal gate-based quantum computer, or a Gaussian Boson Sampling quantum device (Johnson: “the method then creates 50 an objective function and submits that to the quantum annealing computer 60 to solve” [*Examiner note: i.e., the quantum annealing computer is being interpreted as a quantum annealer] [0016]).
Regarding claim 7, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein the digital steps are carried out with the one or more classical computing devices (Cao: “All tests are performed on a computer with Intel(R) Core(TM) i9-7900 X CPU” [Page 629, 3 Experiments]).
Regarding claim 8, Cao in view of Johnson and Choi discloses all of the limitations of claim 7 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein the one or more classical computing devices comprise one or more of: a computer processing unit, a graphics processing unit, and a field-programmable gate array (Cao: “All tests are performed on a computer with Intel(R) Core(TM) i9-7900 X CPU … and two GPUs RTX-2080Ti” [Page 629, 3 Experiments]).
Regarding claim 9, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits, (Choi, Col. 5, Lines 31-47: “The respective bias may control a tunneling rate of each of the physical qubits. The respective bias may control a height of a potential barrier between a first state and a second state of the physical qubit. Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state. Increasing the height of the potential barrier may ensure a state of the physical qubit does not change from the first state to the second state. At least two variables from the Quadratic Unconstrained Binary Optimization problem may be assigned to two respective logical 40 qubit. Each of the respective logical qubits may include at least two of the physical qubits coupled by at least one of the physical qubit couplers. Each pair from the at least two variables may have a relationship in the Quadratic Unconstrained Binary Optimization problem which is represented by a 45 respective controllable coupling between a respective physical qubit from each of the respective logical qubits.”)
the resulting quantum states defining values of binary variables of the quadratic
unconstrained binary optimization problem, (Choi, Col. 5, Lines 38-41: “At least two variables from the Quadratic Unconstrained Binary Optimization problem may be assigned to two respective logical 40 qubit”, Col. 8, Lines 2-4: “Logical qubits may be used to represent one variable from a problem to be solved by the analog computer over several physical qubits.”)
Regarding claim 10, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein solving the quadratic unconstrained binary optimization problem for the first period of time comprises performing further adjustments of the quantum states of the qubits thereby determining resulting quantum states of the qubits; (Choi, Col. 5, Lines 31-47: “The respective bias may control a tunneling rate of each of the physical qubits. The respective bias may control a height of a potential barrier between a first state and a second state of the physical qubit. Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state. Increasing the height of the potential barrier may ensure a state of the physical qubit does not change from the first state to the second state. At least two variables from the Quadratic Unconstrained Binary Optimization problem may be assigned to two respective logical 40 qubit. Each of the respective logical qubits may include at least two of the physical qubits coupled by at least one of the physical qubit couplers. Each pair from the at least two variables may have a relationship in the Quadratic Unconstrained Binary Optimization problem which is represented by a 45 respective controllable coupling between a respective physical qubit from each of the respective logical qubits.”)
the method comprising converting the resulting quantum states of the qubits
into electrical signals; (Choi, Col. 14, Lines 36-43: “In another alternative, NIC 540 may include software and/or hardware that translates signals (representing a solution to a problem or some other form of feedback) from qubit nodes 552 and coupling devices 554. In some cases, analog processor interface module 530 may communicate with driver module 532 rather than directly with NIC 540 in order to send and receive signals from quantum processor 550.”)
wherein inputting the optimal trading trajectories comprises receiving, by the one or more classical computing devices, the electrical signals obtained by converting the resulting quantum states of the qubits; and (Choi, Col. 14, Lines 36-43: “In another alternative, NIC 540 may include software and/or hardware that translates signals (representing a solution to a problem or some other form of feedback) from qubit nodes 552 and coupling devices 554. In some cases, analog processor interface module 530 may communicate with driver module 532 rather than directly with NIC 540 in order to send and receive signals from quantum processor 550.”)
wherein the training of the machine learning algorithm is performed by the one or more classical computing devices (Johnson, ¶[0033]: “The platform library 145 is not necessarily limited to interacting with just quantum processing devices. One or more classical solver libraries 160 for conventional processing devices may also be used by the platform library 145 for various purposes (e.g. to solve some part of a problem that is not well-suited to a quantum processing solution, or to compare an answer obtained on a quantum processing device to an answer obtainable via a classical solver library).”)
Regarding claim 11, Cao discloses an apparatus comprising:
introduce a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets (Cao: “To study a deep learning model for the portfolio optimization problem, we aim at only using the historical stock data during the last M days for training … we represent each input data as a 3D tensor … that includes all stock data (both the volume and the price) of N different tickers during the last M consecutive days” [*Examiner note: the training data is being interpreted as a first set of data, and the last M days are being interpreted as a first period of time] [Page 626, 2.1 Problem Formation]);
and the one or more computing devices being configured to at least further cause the apparatus to: provide a quantum or classical machine learning algorithm that provides a recommended composition of an asset portfolio based on a set of inputs (Cao: “To estimate the output vector w, we consider different deep learning approaches for solving the portfolio optimization… we construct a new ResNet architecture for the problem and create four other combinations of deep neural networks. They are SA + LSTM (Self-Attention model and LSTM), SA + GRU (Self-Attention and GRU)” [*Examiner note: the Self-Attention and GRU deep neural network algorithm is being interpreted as a provided classical machine learning algorithm that solves the portfolio optimization (i.e., provides a recommended composition of an asset portfolio)] [Page 627 – 628, 2.3 Our Proposed Models for the Portfolio Optimization]);
train the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the first period of time and minimizing a predetermined error function for each time unit of the first period of time for which there is historical financial data in the first set of data (Cao: “one can estimate the maximum value of the Sharpe ratio by solving the following optimization problem … one can use the stochastic gradient descent method for approximating the optimal w” [*Examiner note: i.e., w is being interpreted as the optimal trading trajectories] [Page 627, 2.2 A New Loss Function for the Sharpe-Ratio Maximization] … “we apply ResNet for estimating the optimal value for the vector w in the loss function” [*Examiner note: i.e., estimating the optimal value for a loss function is being interpreted as minimizing a predetermined error function for w (i.e., input optimal trading trajectories)] [Page 628, 2.3 Our Proposed Models for the Portfolio Optimization]);
introduce a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets (Cao: “On each day, we collect the information of both “price” and “volume” of these 381 tickers and y, the daily return on the market in the next K days” [*Examiner note: i.e., the daily return on the market for the next K days is being interpreted as a second set of data for a second period of time] [Page 631, 3.2 Data Preparation]);
and provide a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein (Cao: “The main goal in our work is to estimate the optimal solution wopt for the portfolio optimization problem (3) in order to obtain the maximum Sharpe ratio during the next K days … we aim at only using the historical stock data during the last M days as training and then predict the optimal equally weighted portfolio having the highest Sharpe ratio during the next K days.” [Page 626, 2.1 Problem Formation]).
Cao fails to disclose:
a quantum device;
the one or more computing devices comprising one or more classical computing devices; and a signal converter
and one or more computing devices communicatively coupled with the quantum device; the one or more computing devices being configured to at least cause the apparatus to:
provide a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio;
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced;
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device;
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced;
the quantum device, with the adjusted quantum states of the qubits, being configured to at least cause the apparatus to solve the quadratic unconstrained binary optimization problem for the first period of time, thereby obtaining optimal trading trajectories for the first period of time;
However, Johnson discloses:
a quantum device (Johnson: “FIG. 1 is a diagram illustrating a financial portfolio optimization method implemented using a quantum processing device” [0006]);
by one or more classical computing devices (Johnson, ¶[0033]: “The platform library 145 is not necessarily limited to interacting with just quantum processing devices. One or more classical solver libraries 160 for conventional processing devices may also be used by the platform library 145 for various purposes (e.g. to solve some part of a problem that is not well-suited to a quantum processing solution, or to compare an answer obtained on a quantum processing device to an answer obtainable via a classical solver library).”)
and one or more computing devices communicatively coupled with the quantum device; the one or more computing devices being configured to at least cause the apparatus to (Johnson: “The platform library 145 is not necessarily limited to interacting with just quantum processing devices. One or more classical solver libraries 160 for conventional processing devices may also be used by the platform library 145 for various purposes (e.g. to solve some part of a problem that is not well-suited to a quantum processing solution, or to compare an answer obtained on a quantum processing device to an answer obtainable via a classical solver library).” [0033]):
provide a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio (Johnson: “An objective function is formulated based on the list of lots and on the budget. The objective function has a form suitable for solution by a quantum processing device, which is used to optimize the objective function” [Abstract] … “The third term is based on a component Gcost for the cost or budget allocated to purchase assets.” [*Examiner note: i.e., the objective function contains a cost function Gcost] [0020] … “Module 220 uses heuristics to convert a higher-order polynomial binary optimization problem into a quadratic binary optimization problem [*Examiner note: i.e., the objective function is a quadratic unconstrained binary optimization problem] [0047]);
the quantum device, with the adjusted quantum states of the qubits, being configured to at least cause the apparatus to solve the quadratic unconstrained binary optimization problem for the first period of time, thereby obtaining optimal trading trajectories for the first period of time (Johnson: “Given the financial offerings 10, the method runs on a digital computer 20 and queries the market to download historical market data 30 to computer the daily rate of return, its variance, and its covariance over a period of time specified by the portfolio manager” [*Examiner note: emphasis added] [0014] … “Once the portfolio manager chooses 40 his targeted level of expected rate of return for the entire portfolio, the objective function 50 can be reconstructed to optimize for a desired expected rate of return and minimum variance. The problem is resubmitted to the quantum annealing computer 60 to determine the selection of lots to purchase that meet the budget, expected rate of return, and minimum variance in price” [0016] … “The result 70 is output to the portfolio manager showing a spectrum of solutions including the optimum along with a number of near-optimum solutions for consideration.” [*Examiner note: i.e., returning the optimum solution is solving the objective (i.e., quadratic unconstrained binary optimization) function to obtain optimal trading trajectories] [0017]);
Cao and Johnson both disclose inventions relating to the field of endeavor relating to portfolio optimization and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to input the optimal trading trajectories generated by a quantum device as taught by Johnson into the training procedure taught by Cao. One having ordinary skill in the art would have been motivated to make this change before the effective filing date of the claimed invention because quantum devices are faster than traditional computer for portfolio optimization (Johnson: “This method formulates the portfolio optimization problem and solves for the best solutions using the power of quantum annealing computers to achieve solutions much faster than existing methods running on traditional digital computer hardware alone” [0012]).
However, Choi explicitly discloses:
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced; (Choi, Col. 14, Lines 40-43: “In some cases, analog processor interface module 530 may communicate with driver module 532 rather than directly with NIC 540 in order to send and receive signals from quantum processor 550.”)
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device; (Choi, Col. 14, Lines 26-36: “Where computing system 500 includes a driver module 532, the driver module 532 may include instructions to output signals to quantum processor 550. NIC 540 may include appropriate hardware required for interfacing with qubit nodes 552 and coupling devices 554, either directly or through readout device 556, qubit control system 558, and/or coupling device control system 560. Alternatively, NIC 540 may include software and/or hardware that translate commands from driver module 532 into signals (e.g., voltages, currents, optical signals, etc.) that are directly applied to qubit nodes 552 and coupling devices 554.”)
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced; (Choi, Col. 5, Lines 31-47: “The respective bias may control a tunneling rate of each of the physical qubits. The respective bias may control a height of a potential barrier between a first state and a second state of the physical qubit. Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state. Increasing the height of the potential barrier may ensure a state of the physical qubit does not change from the first state to the second state. At least two variables from the Quadratic Unconstrained Binary Optimization problem may be assigned to two respective logical 40 qubit. Each of the respective logical qubits may include at least two of the physical qubits coupled by at least one of the physical qubit couplers. Each pair from the at least two variables may have a relationship in the Quadratic Unconstrained Binary Optimization problem which is represented by a 45 respective controllable coupling between a respective physical qubit from each of the respective logical qubits.”)
with the adjusted quantum states of the qubits (Choi, Col. 5, Lines 34-36: “Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state.”)
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Cao and Choi. Cao teaches a method for the optimal trajectories approximated. Choi teaches methods for analog processing using quantum computing devices to solve Quadratic Unconstrained Binary Optimization problems. One of ordinary skill would have motivation to combine Cao and Choi to enable direct physical interaction with the quantum hardware, allowing qubit states to be tuned according to the encoded problem parameters.
Regarding claim 12, Cao in view of Johnson and Choi discloses all of the limitations of claim 11 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein: the one or more computing devices are configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, introduce a third set of data into the problem, the third set of data comprising historical financial data for the second period of time (Cao: “we use the time windows of M consecutive days for extracting the input data of proposed models. On each day, we collect the information of both “price” and “volume“ of these 381 tickers and y, the daily return on the market in the next K days (K = 19). Consequently, the input data has the shape (381, 64, 2) and we move the time window during the studying period of time (from January 1, 2013, to July 31, 2019) to obtain 1415 samples.” [*Examiner note: moving the time window from the first M days forward to the next M days would encompass the next K days from the previous iteration, i.e., this third data set would be the historical data for the next K days (i.e., the second period of time)] [Page 631, 3.2 Data Preparation]);
the quantum device is configured to at least further cause the apparatus to, after historical financial data is available for the second period of time, solve the quadratic unconstrained binary optimization problem for the second period of time, thereby obtaining optimal trading trajectories for the second period of time (Johnson: “Given the financial offerings 10, the method runs on a digital computer 20 and queries the market to download historical financial data 30 to compute the daily rate of return, its variance, and its covariance over a period of time specified by the portfolio manager.” [*Examiner note: i.e., the manager can specify the second period of time] [0014] … “The problem is resubmitted to the quantum annealing computer 60 to determine the selection of lots to purchase that meet the budget, expected rate of return, and minimum variance in price” [0016] … “The result 70 is output to the portfolio manager showing a spectrum of solutions including the optimum along with a number of near-optimum solutions for consideration.” [*Examiner note: i.e., returning the optimum solution is solving the objective (i.e., quadratic unconstrained binary optimization) function to obtain optimal trading trajectories] [0017]);
and the one or more computing devices are configured to at least further cause the apparatus to, after historical financial data is available for the second period of time: train the machine learning algorithm by both inputting the optimal trading trajectories obtained by the quantum device for the second period of time and minimizing a predetermined error function for each time unit of the second period of time for which there is historical financial data in the third set of data (Cao: “only using the historical stock data during the last M days for training or updating the proposed model, and then doing prediction for the equally weighted portfolio having the highest Sharpe ratio during the next K days” [*Examiner note: emphasis added. i.e., after the algorithm has been trained on data from the first window, it is updated on data from the following windows] [Page 625, 1 Introduction]);
introduce a fourth set of data into the machine learning algorithm, the fourth set of data comprising financial data for a third period of time that is posterior to the second period of time (Cao: “The main goal in our work is to estimate the optimal solution wopt for the portfolio optimization problem (3) in order to obtain the maximum Sharpe ratio during the next K days … we aim at only using the historical stock data during the last M days as training and then predict the optimal equally weighted portfolio having the highest Sharpe ratio during the next K days.” [Page 626, 2.1 Problem Formation] … we move the time window during the studying period of time (from January 1, 2013, to July 31, 2019) to obtain 1415 samples.” [*Examiner note: these windows include a fourth set of data that comprises financial data for a third period of time posterior to the second period of time] [Page 631, 3.2 Data Preparation]);
and provide a recommended portfolio composition for the third period of time by running the trained machine learning algorithm with the fourth set of data introduced therein (Cao: “The main goal in our work is to estimate the optimal solution wopt for the portfolio optimization problem (3) in order to obtain the maximum Sharpe ratio during the next K days … we aim at only using the historical stock data during the last M days as training and then predict the optimal equally weighted portfolio having the highest Sharpe ratio during the next K days.” [Page 626, 2.1 Problem Formation]).
Regarding claim 14, Cao in view of Johnson and Choi discloses all of the limitations of claim 11 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein the machine learning algorithm comprises a neural network or a variational quantum circuit (Cao: “by implementing an appropriate deep neural network to estimate the optimal solution …, we can derive the final output vector wopt” [Page 627, 2.2 A New Loss Function for the Sharpe-Ratio Maximization]).
Regarding claim 16, Cao in view of Johnson discloses all of the limitations of claim 11 as shown in the rejection above. Cao in view of Johnson also discloses:
wherein the quantum device comprises one of: a quantum annealer, a hybrid quantum-classical machine, a universal gate-based quantum computer, or a Gaussian Boson Sampling quantum device (Johnson: “the method then creates 50 an objective function and submits that to the quantum annealing computer 60 to solve” [*Examiner note: i.e., the quantum annealing computer is being interpreted as a quantum annealer] [0016]).
Regarding claim 17, Cao in view of Johnson and Choi discloses all of the limitations of claim 11 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein the one or more classical computing devices comprise one or more of: a computer processing unit, a graphics processing unit, and a field-programmable gate array (Cao: “All tests are performed on a computer with Intel(R) Core(TM) i9-7900 X CPU … and two GPUs RTX-2080Ti” [Page 629, 3 Experiments]).
Regarding claim 18, Cao in view of Johnson and Choi discloses all of the limitations of claim 11 as shown in the rejection above. Cao in view of Johnson and Choi also discloses:
wherein the first period of time comprises a plurality of days and the second period of time comprises one day (Cao: “using the historical stock data during the last M days for training or updating the proposed model, and then doing prediction for the equally weighted portfolio having the highest Sharpe ratio during the next K days” [*Examiner note: i.e., the last M days comprises a plurality days, and if K = 1 then the second time period comprises one day] [Page 625, 1 Introduction]).
Regarding claim 20, Cao discloses:
A non-transitory computer-readable medium encoded with instructions that, when executed by at least one processor or hardware, make an apparatus to at least perform the following (Cao: “All tests are performed on a computer with Intel(R) Core(TM) i9-7900X CPU” [Page 629, 3 Experiments]):
Introducing, by one or more classical computing devices, a first set of data into the problem, the first set of data comprising historical financial data for a first period of time, the historical financial data at least comprising prices of considered assets (Cao: “To study a deep learning model for the portfolio optimization problem, we aim at only using the historical stock data during the last M days for training … we represent each input data as a 3D tensor … that includes all stock data (both the volume and the price) of N different tickers during the last M consecutive days” [*Examiner note: the training data is being interpreted as a first set of data, and the last M days are being interpreted as a first period of time] [Page 626, 2.1 Problem Formation]);
providing a quantum or classical machine learning algorithm that provides a recommended composition of an asset portfolio based on a set of inputs (Cao: “To estimate the output vector w, we consider different deep learning approaches for solving the portfolio optimization… we construct a new ResNet architecture for the problem and create four other combinations of deep neural networks. They are SA + LSTM (Self-Attention model and LSTM), SA + GRU (Self-Attention and GRU)” [*Examiner note: the Self-Attention and GRU deep neural network algorithm is being interpreted as a provided classical machine learning algorithm that solves the portfolio optimization (i.e., provides a recommended composition of an asset portfolio)] [Page 627 – 628, 2.3 Our Proposed Models for the Portfolio Optimization]);
training the machine learning algorithm by both inputting optimal trading trajectories outputted by the [quantum] device for the first period of time and minimizing a predetermined error function for each time unit of the first period of time for which there is historical financial data in the first set of data (Cao: “we apply ResNet for estimating the optimal value for the vector w in the loss function” [*Examiner note: i.e., estimating the optimal value for a loss function is being interpreted as minimizing a predetermined error function] [Page 628, 2.3 Our Proposed Models for the Portfolio Optimization]);
introducing a second set of data into the machine learning algorithm, the second set of data comprising financial data for a second period of time that is posterior to the first period of time, the financial data at least comprising prices of the considered assets (Cao: “On each day, we collect the information of both “price” and “volume” of these 381 tickers and y, the daily return on the market in the next K days” [*Examiner note: i.e., the daily return on the market for the next K days is being interpreted as a second set of data for a second period of time] [Page 631, 3.2 Data Preparation]);
and providing a recommended portfolio composition for the second period of time by running the trained machine learning algorithm with the second set of data introduced therein (Cao: “The main goal in our work is to estimate the optimal solution wopt for the portfolio optimization problem (3) in order to obtain the maximum Sharpe ratio during the next K days … we aim at only using the historical stock data during the last M days as training and then predict the optimal equally weighted portfolio having the highest Sharpe ratio during the next K days.” [Page 626, 2.1 Problem Formation]).
Cao fails to disclose:
That the device is a quantum device
providing a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio;
by one or more classical computing devices
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced;
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device;
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced;
However, Johnson discloses:
That the device is a quantum device (Johnson: “FIG. 1 is a diagram illustrating a financial portfolio optimization method implemented using a quantum processing device” [0006])
by one or more classical computing devices (Johnson, ¶[0033]: “The platform library 145 is not necessarily limited to interacting with just quantum processing devices. One or more classical solver libraries 160 for conventional processing devices may also be used by the platform library 145 for various purposes (e.g. to solve some part of a problem that is not well-suited to a quantum processing solution, or to compare an answer obtained on a quantum processing device to an answer obtainable via a classical solver library).”)
providing a quadratic unconstrained binary optimization problem defined by an equation with a cost function for optimization of trading trajectories of an asset portfolio (Johnson: “An objective function is formulated based on the list of lots and on the budget. The objective function has a form suitable for solution by a quantum processing device, which is used to optimize the objective function” [Abstract] … “The third term is based on a component Gcost for the cost or budget allocated to purchase assets.” [*Examiner note: i.e., the objective function contains a cost function Gcost] [0020] … “Module 220 uses heuristics to convert a higher-order polynomial binary optimization problem into a quadratic binary optimization problem [*Examiner note: i.e., the objective function is a quadratic unconstrained binary optimization problem] [0047]);
Cao and Johnson both disclose inventions relating to the field of endeavor relating to portfolio optimization and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to input the optimal trading trajectories generated by a quantum device as taught by Johnson into the training procedure taught by Cao. One having ordinary skill in the art would have been motivated to make this change before the effective filing date of the claimed invention because quantum devices are faster than traditional computer for portfolio optimization (Johnson: “This method formulates the portfolio optimization problem and solves for the best solutions using the power of quantum annealing computers to achieve solutions much faster than existing methods running on traditional digital computer hardware alone” [0012]).
However, Choi explicitly discloses:
thereby obtaining electrical signals of the one or more classical computing devices, the electrical signals defining the problem with the first set of data introduced; (Choi, Col. 14, Lines 40-43: “In some cases, analog processor interface module 530 may communicate with driver module 532 rather than directly with NIC 540 in order to send and receive signals from quantum processor 550.”)
converting the electrical signals into another type of signals, the another type of signals being to adjust quantum states of qubits of a quantum device; (Choi, Col. 14, Lines 26-36: “Where computing system 500 includes a driver module 532, the driver module 532 may include instructions to output signals to quantum processor 550. NIC 540 may include appropriate hardware required for interfacing with qubit nodes 552 and coupling devices 554, either directly or through readout device 556, qubit control system 558, and/or coupling device control system 560. Alternatively, NIC 540 may include software and/or hardware that translate commands from driver module 532 into signals (e.g., voltages, currents, optical signals, etc.) that are directly applied to qubit nodes 552 and coupling devices 554.”)
adjusting, by interaction between the another type of signals and the qubits, quantum states of the qubits for solving the problem with the first set of data introduced; (Choi, Col. 5, Lines 31-47: “The respective bias may control a tunneling rate of each of the physical qubits. The respective bias may control a height of a potential barrier between a first state and a second state of the physical qubit. Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state. Increasing the height of the potential barrier may ensure a state of the physical qubit does not change from the first state to the second state. At least two variables from the Quadratic Unconstrained Binary Optimization problem may be assigned to two respective logical 40 qubit. Each of the respective logical qubits may include at least two of the physical qubits coupled by at least one of the physical qubit couplers. Each pair from the at least two variables may have a relationship in the Quadratic Unconstrained Binary Optimization problem which is represented by a 45 respective controllable coupling between a respective physical qubit from each of the respective logical qubits.”)
with the adjusted quantum states of the qubits (Choi, Col. 5, Lines 34-36: “Decreasing the height of the potential barrier may allow a state of the physical qubit to change 35 from the first state to the second state.”)
It would have obvious to one of ordinary skill in the art before the effective filing date of the present application to combine Cao and Choi. Cao teaches a method for the optimal trajectories approximated. Choi teaches methods for analog processing using quantum computing devices to solve Quadratic Unconstrained Binary Optimization problems. One of ordinary skill would have motivation to combine Cao and Choi to enable direct physical interaction with the quantum hardware, allowing qubit states to be tuned according to the encoded problem parameters.
Claims 3 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Cao in view of Johnson, Choi and in further view of Nazari et al. (US PGPUB No US2019/0295169 A1) (hereafter referred to as Nazari)
Regarding claim 3, Cao in view of Johnson, Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi fails to disclose:
further comprising digitally commanding making one or more investments based on the recommended portfolio composition provided.
However, Nazari discloses:
further comprising digitally commanding making one or more investments based on the recommended portfolio composition provided (Nazari: “This framework implements several algorithms to quantify the uncertainty in the predictions of future assets values, and provides to the user important hints and advice to optimize entries and exits in a semi-automatic mode. The framework also has the capability of performing automatic trading once the personal risk profile of the trader has been defined.” [0028] … “The method also has the option of the automatic mode to trade the selected stocks without the user decision.” [0053]).
Cao in view of Johnson, Choi and Nazari both disclose inventions relating to the field of endeavor relating to portfolio optimization and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to have the model taught by Cao in view of Johnson and Choi automatically make the one or more investments as taught by Nazari rather than having a human portfolio manager make the investments. One having ordinary skill in the art would have been motivated to make this change because the model is not susceptible to human error or emotion (Nazari: “This automatic mode has the advantage of avoiding fear and greed, since it is based purely in quantitative analysis” [0053]).
Regarding claim 13, Cao in view of Johnson and Choi discloses all of the limitations of claim 11 as shown in the rejection above. Cao in view of Johnson and Choi fails to disclose:
wherein the one or more computing devices are configured to at least further cause the apparatus to command making one or more investments based on the recommended portfolio composition provided.
However, Nazari discloses:
wherein the one or more computing devices are configured to at least further cause the apparatus to command making one or more investments based on the recommended portfolio composition provided (Nazari: “This framework implements several algorithms to quantify the uncertainty in the predictions of future assets values, and provides to the user important hints and advice to optimize entries and exits in a semi-automatic mode. The framework also has the capability of performing automatic trading once the personal risk profile of the trader has been defined.” [0028] … “The method also has the option of the automatic mode to trade the selected stocks without the user decision.” [0053]).
Cao in view of Johnson, Choi and Nazari both disclose inventions relating to the field of endeavor relating to portfolio optimization and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to have the model taught by Cao in view of Johnson and Choi automatically make the one or more investments as taught by Nazari rather than having a human portfolio manager make the investments. One having ordinary skill in the art would have been motivated to make this change because the model is not susceptible to human error or emotion (Nazari: “This automatic mode has the advantage of avoiding fear and greed, since it is based purely in quantitative analysis” [0053]).
Claims 5 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Cao in view of Johnson, Choi and in further view of Orus et al., (Quantum computing for finance: Overview and prospects) (hereafter referred to as Orus).
Regarding claim 5, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi fails to disclose:
wherein the cost function is
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A at least comprises the following terms
PNG
media_image2.png
55
383
media_image2.png
Greyscale
or at least comprises the following terms
PNG
media_image3.png
47
534
media_image3.png
Greyscale
, where wt is a vector of the components of which are the percentages of each asset in the portfolio at time t, μt is a vector of expected returns at time t, γ is a parameter controlling the volatility of the portfolio, ∑t is a matrix of covariances of the returns at time t, v-t is a percentage of transaction costs, ∆wt is a change in the composition of the vector of assets between time t and time t + 1, Λt is a matrix of market impact at time t, and ti and tf are an initial time and a final time of a respective period of time.
However, Orus discloses:
wherein the cost function is
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A at least comprises the following terms
PNG
media_image2.png
55
383
media_image2.png
Greyscale
or at least comprises the following terms
PNG
media_image3.png
47
534
media_image3.png
Greyscale
, where wt is a vector of the components of which are the percentages of each asset in the portfolio at time t, μt is a vector of expected returns at time t, γ is a parameter controlling the volatility of the portfolio, ∑t is a matrix of covariances of the returns at time t, v-t is a percentage of transaction costs, ∆wt is a change in the composition of the vector of assets between time t and time t + 1, Λt is a matrix of market impact at time t, and ti and tf are an initial time and a final time of a respective period of time (Orus: “Let us consider the problem of dynamic portfolio optimization … The cost function which was optimized was the return:
PNG
media_image4.png
103
620
media_image4.png
Greyscale
with μ the forecast returns, w the holdings, ∑ the forecast covariance tensor, and γ the risk aversion. The third and fourth terms represent different contributions to transaction costs” [*Examiner note: i.e., this cost function is being interpreted as equivalent to
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A comprises at least the term
PNG
media_image2.png
55
383
media_image2.png
Greyscale
] [Page 6, 3.1. Optimal trading trajectory]).
Cao in view of Johnson, Choi and Orus both disclose inventions relating to the field of endeavor of portfolio optimization and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to replace the cost function taught by Cao in view of Johnson, Choi with the cost function disclosed by Orus. One having ordinary skill in the art would have been motivated to make this change because this cost function is immediately usable by a quantum annealer (Orus: “It was suggested in Ref. [48] that the discrete multi-period version of this problem was amenable to quantum annealers. This idea was implemented on a D-wave quantum processor” [Page 6, 3.1. Optimal trading trajectory]).
Regarding claim 15, Cao in view of Johnson and Choi discloses all of the limitations of claim 1 as shown in the rejection above. Cao in view of Johnson and Choi fails to disclose:
wherein the cost function is
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A at least comprises the following terms
PNG
media_image2.png
55
383
media_image2.png
Greyscale
or at least comprises the following terms
PNG
media_image3.png
47
534
media_image3.png
Greyscale
, where wt is a vector of the components of which are the percentages of each asset in the portfolio at time t, μt is a vector of expected returns at time t, γ is a parameter controlling the volatility of the portfolio, ∑t is a matrix of covariances of the returns at time t, v-t is a percentage of transaction costs, ∆wt is a change in the composition of the vector of assets between time t and time t + 1, Λt is a matrix of market impact at time t, and ti and tf are an initial time and a final time of a respective period of time.
However, Orus discloses:
wherein the cost function is
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A at least comprises the following terms
PNG
media_image2.png
55
383
media_image2.png
Greyscale
or at least comprises the following terms
PNG
media_image3.png
47
534
media_image3.png
Greyscale
, where wt is a vector of the components of which are the percentages of each asset in the portfolio at time t, μt is a vector of expected returns at time t, γ is a parameter controlling the volatility of the portfolio, ∑t is a matrix of covariances of the returns at time t, v-t is a percentage of transaction costs, ∆wt is a change in the composition of the vector of assets between time t and time t + 1, Λt is a matrix of market impact at time t, and ti and tf are an initial time and a final time of a respective period of time (Orus: “Let us consider the problem of dynamic portfolio optimization … The cost function which was optimized was the return:
PNG
media_image4.png
103
620
media_image4.png
Greyscale
with μ the forecast returns, w the holdings, ∑ the forecast covariance tensor, and γ the risk aversion. The third and fourth terms represent different contributions to transaction costs” [*Examiner note: i.e., this cost function is being interpreted as equivalent to
PNG
media_image1.png
59
144
media_image1.png
Greyscale
, where A comprises at least the term
PNG
media_image2.png
55
383
media_image2.png
Greyscale
] [Page 6, 3.1. Optimal trading trajectory]).
Cao in view of Johnson, Choi and Orus both disclose inventions relating to the field of endeavor of portfolio optimization and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to replace the cost function taught by Cao in view of Johnson and Choi with the cost function disclosed by Orus. One having ordinary skill in the art would have been motivated to make this change because this cost function is immediately usable by a quantum annealer (Orus: “It was suggested in Ref. [48] that the discrete multi-period version of this problem was amenable to quantum annealers. This idea was implemented on a D-wave quantum processor” [Page 6, 3.1. Optimal trading trajectory]).
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over Cao in view of Johnson, Choi and in further view of Macrooption (How to Download Historical Data from Yahoo Finance) (hereafter referred to as Macrooption).
Regarding claim 19, Cao in view of Johnson, Choi discloses all of the limitations of claim 18 as shown in the rejection above. Cao in view of Johnson and Choi fails to disclose:
wherein the one day of the second period of time is today or yesterday.
However, Macrooption discloses:
wherein the one day of the second time period is today or yesterday (Macrooption: “you can adjust the date range, data type (usually you want Historical Prices, which is set by default) and frequency (you probably want Daily, set by default). Don’t forget to click Apply if you’ve made any changes. Then click Download Data” [*Examiner note: this tool can download data from today or yesterday] [Page 3, Preparing the Historical Data]).
Cao in view of Johnson, Choi and Macrooption both disclose inventions relating to operating on financial market data and are therefore analogous. It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to use the teachings of Macrooption to gather today’s market data, and then use that as the next day taught by Cao in view of Johnson and Choi. One having ordinary skill in the art would have been motivated to make this change before the effective filing date of the claimed invention because it allows the data to be downloaded in a format that can be easily saved or opened immediately (Macrooption: “The website will offer a CSV file, usually named table.csv, which you can either save toyour computer or immediately open.” [Page 4, Preparing the Historical Data]).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AMY TRAN whose telephone number is (571)270-0693. The examiner can normally be reached Monday - Friday 7:30 am - 5:00 pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, David Yi can be reached at (571) 270-7519. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/AMY TRAN/Examiner, Art Unit 2126
/VAN C MANG/Primary Examiner, Art Unit 2126