NON-FINAL REJECTION, FIRST DETAILED ACTION
Status of Prosecution
The present application, 18/511,994 filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
The application, filed in the Office on Nov. 16, 2023.
Claims 1-20 are pending and are all rejected. Claims 1 and 20 are independent.
Status of Claims
Claim 14 is objected to.
Claims 8 and 13 are rejected as indefinite under 25 USC § 112(b).
Claims 1, 5-7, 9, 11-12, 14-15, 17-18 and 20 are rejected under 35 U.S.C. § 103 as being unpatentable over non-patent literature Pfeffer et al. (“Pfeffer”), “Hybrid quantum-classical reservoir computing of thermal convection flow” published in 2022 in view of non-patent literature Govia et al. (“Govia”), “Nonlinear input transformations are ubiquitous in quantum reservoir computing,” published in 2022.
Claims 2-4, 10 and 19 are rejected under 35 U.S.C. § 103 as being unpatentable over Pfeffer in view of Govia in further of view non-patent literature Fry et al. (“Fry”), “Optimizing quantum noise‑induced reservoir computing for nonlinear and chaotic time series prediction,” published in 2023.
Claim 8 is rejected under 35 U.S.C. § 103 as being unpatentable over Pfeffer in view of Govia in further of view non-patent literature Bogris et al. (“Bogris”), “Fabry-Perot Lasers as Enablers for Parallel Reservoir Computing,” published in 2021.
Claims 16 is rejected under 35 U.S.C. § 103 as being unpatentable over Pfeffer in view of Govia in further of view non-patent literature Strock et al. (“Strock”), “A Simple Reservoir Model of Working Memory with Real Values” published in 2018.
Objection
Claim 14 is objected to for what appears to either be a typographical error or a misuse of a common term. “Highly dimensional Hilbert space,” is recited and Examiner believes the intention was “high dimensional Hilbert space.” A response and appropriate amendment if needed s requested. For purposes of compact prosecution in the instant action, the claim will be construed as “high dimensional Hilbert space.”
Claim Interpretation – 112(f)
The following is a quotation of 35 U.S.C. § 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. § 112(f) because the claim limitations use a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitations are in claim 20:
data input and transformation module;
quantum data encoding module;
a quantum measurement model;
a quantum feedback module;
a random transformation module; and
a reservoir state creation module.
Because these claim limitations are being interpreted under 35 U.S.C. § 112(f) they are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have these limitations interpreted under 35 U.S.C. § 112(f), Applicant may: (1) amend the claim limitations to avoid them being interpreted under 35 U.S.C. § 112(f) (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitations recite sufficient structure to perform the claimed function so as to avoid them being interpreted under 35 U.S.C. § 112(f).
Claim Rejections – 35 USC § 112
The following is a quotation of 35 U.S.C. § 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claim 8 is rejected as indefinite. The term “Fourier-like” in claim 8 is a relative term which renders the claim indefinite. The term “Fourier-like” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. Correction or clarification is requested.
Claim 13 is rejected as being indefinite, due to the great deal of confusion and uncertainty as to the proper interpretation of the limitations of the claim. Specifically, Claim 13 recites, “wherein the updated measurement vector further comprises single-qubit expectation values and multi-qubit correlators, wherein both the single-qubit expectation values and the multi-qubit correlators are defined on a measurement graph.” A review of the Specification and consideration of the state of the art does not yield what may be a “measurement graph.”
No prior art rejection is made for claim 13. See MPEP 2173.06(II) (“As stated in In re Steele, 305 F.2d 859, 134 USPQ 292 (CCPA 1962), a rejection under 35 U.S.C. § 103 should not be based on considerable speculation about the meaning of terms employed in a claim or assumptions that must be made as to the scope of the claims.”).
Applicant’s is invited to interview with Examiner to discuss this claim.
Claim Rejections – 35 USC § 103
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.
A.
Claims 1, 5-7, 9, 11-12, 14-15, 17-18 and 20 are rejected under 35 U.S.C. § 103 as being unpatentable over non-patent literature Pfeffer et al. (“Pfeffer”), “Hybrid quantum-classical reservoir computing of thermal convection flow” published in 2022 in view of non-patent literature Govia et al. (“Govia”), “Nonlinear input transformations are ubiquitous in quantum reservoir computing,” published in 2022.
As to Claim 1, Pfeffer teaches: A hybrid quantum-classical computing method comprising:
receiving time dependent input data of a dynamical system (Pfeffer: Sec. IV.A., a time series A4(t) is received);
encoding the input data into a quantum circuit by performing a first set of quantum operations, wherein the quantum circuit comprises a plurality of layers, wherein the layers include data encoding layers, and a random transformation layer (Pfeffer: Fig. 2(b), eq. 13 denotes the updating of the dynamical part in three blocks of unitary linear transformation, as noted in the figure, which involves encoding and random transformations by the random rotation angles β);
PNG
media_image1.png
55
569
media_image1.png
Greyscale
encoding a measurement feedback from a previous measurement vector into the quantum circuit by performing a second set of quantum operations (Pfeffer: Fig. 2b, Sec. III A, equation 13, the updating is performed by three unitary linear transformations including past system state vector xt);
operating the quantum circuit to generate an updated measurement vector (Pfeffer: Sec. III.B, “Figure 2(b) shows the corresponding circuit diagram of the quantum reservoir which consists of three circuit blocks as lined out in Eq. (13)”);
generating a reservoir state based on the updated measurement vector, a previous reservoir state and the input data (Pfeffer: eq. 13, “The reservoir state evolves from time t to t + ∆t with a fixed time step width ∆t” as follows in the equation, which includes each of those block elements).
PNG
media_image2.png
262
807
media_image2.png
Greyscale
Pfeffer may not explicitly teach: transforming a first element of the input data using a plurality of transformation matrices to obtain a set of transformed data;
encoding the transformed data into a quantum circuit by performing a first set of quantum operations, wherein the quantum circuit comprises a plurality of layers, wherein the layers include data encoding layers, and a random transformation layer;
encoding the transformed data into a quantum circuit by performing a first set of quantum operations, wherein the quantum circuit comprises a plurality of layers, wherein the layers include data encoding layers, feedback layers, and a random transformation layer.
Govia teaches in general concepts related to defining a conceptual framework for separating constituent components and determining their impacts on performance (Govia: Abstract). Specifically, Govia teaches that input signals are initialized and pre-processed by using an input mask matrix (Govia: Sec. 2, eq. 1, the matrix Win is used).
PNG
media_image3.png
51
865
media_image3.png
Greyscale
Subsequently, the initialization leads to the input encoding and internal evolution and measurement phases in the discrete version of quantum reservoir computing (Govia: Fig. 1).
PNG
media_image4.png
881
899
media_image4.png
Greyscale
It would have been obvious to a person having ordinary skill in the art at a time before the effective filing date of the application to have modified the Pfeffer disclosures and teachings by pre processing the input signals with matrices as taught and suggested by Govia. Such a person would have been motivated to do so with a reasonable expectation of success to either denoise or properly adjust the input signals for optimal performance via the masking.
As to Claim 5, Pfeffer and Govia teach the limitations of claim 1.
Govia further teaches: wherein the method further comprises using the quantum circuit and a trained classical processing module to generate predictions of a future state of the dynamical system, wherein the classical processing module receives as input non-linear transformations of the reservoir state and non-linear transformations of the time dependent input data (Govia: Sec. 3, “By definition, a linear encoding in one basis implies a linear encoding in all other bases, since they are connected by linear transformations. Note that the other terms of equations (3) and (4), encapsulated in ‘+· · ·’, can include nonlinear functions of the previous state of the reservoir nodes, and therefore of the input history, i.e. ul for l < j or u(s) for s < t.)”
As to Claim 6, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: wherein the method further comprises initiating entanglement among a plurality of qubits of the quantum circuit in the data encoding layers (Pfeffer: Introduction, “The strong encoding capabilities of fully entangled quantum reservoirs are demonstrated in the present flow case by runs with qubit numbers n < Ndof “).
As to Claim 7, Pfeffer and Govia teach the limitations of claim 1.
Govia further teaches: wherein each of the plurality of transformation matrices comprise a plurality of fixed transformation matrices, which remain invariant during both training and prediction phases of the computing process (Examine asserts the input mask may be fixed and not changed as it would be used for fixed reservoir dynamics).
As to Claim 9, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: wherein each of the first and second set of quantum operations comprise one or a combination of one or more of: single-qubit rotations around at least one of X, Y and Z-axes; Controlled-Phase gate operation; or fSim gate operation; or 2-qubit XY rotation (Pfeffer: Introduction, “The quantum reservoir is composed of a sequence of elementary single and two-qubit quantum gates which form a complex quantum circuit.”; Sec. III. B, “This is done by rotation gates Rγ (4π ptk )”).
As to Claim 11, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: wherein the reservoir circuit layer includes reservoir units corresponding to a set of quantum gates, wherein parameters of the quantum gates are independent of measurement feedback, input data, or reservoir state (Pfeffer: Introduction, “The quantum reservoir is composed of a sequence of elementary single and two-qubit quantum gates which form a complex quantum circuit.”; Examiner notes that the gates appear to be independent of the measurement feedback, input data, or reservoir state, as the initialization of the reservoir parameters are completed initially and the readout is optimized and changed; not the initial parameters).
As to Claim 12, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: wherein the generation of the reservoir state is based on a plurality of activation functions, which introduce non-linearities into the operations performed by the quantum circuit, and a leak rate parameter (Pfeffer: eqs. 14 and 15 contains a leaking rate that tempters the update of the reservoir state; sec. III.A., “The nonlinearity is connected to the classical data loading as will be discussed in the next subsection. Equation (14) contains a leaking rate 0 _ ε _ 1 that blends both terms. In the classical reservoir computing model, the update of the reservoir state ψt c would be given by [eq. 15].”).
As to Claim 14, Pfeffer and Govia teach the limitations of claim 1.
Govia further teaches: wherein the method utilizes properties of a highly dimensional Hilbert space as a reservoir for encoding chaotic dynamics into the quantum circuit (Govia: Sec. 5, infitite dimensional Hilbert spaces are used for QRC’s known as continuous variable systems).
As to Claim 15, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: wherein prior measurements corresponding to the measurement feedback are derived from the quantum circuit parameterized by a preceding iteration of the computing method, wherein the prior measurements parameterize feedback layers (Pfeffer: Fig. 2b, Sec. III A, equation 13, the updating is performed by three unitary linear transformations including past system state vector xt; eq. 14 also gives an update rule by iterations).
As to Claim 17, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: wherein the time dependent input data comprises multicomponent time series data vectors (Pfeffer: Sec. IV.A, the Lorenz-type model is a multidimenionsal model with time series as a component).
As to Claim 18, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: further comprising using a classical co-processor to receive classical output from the quantum circuit and apply post-processing to produce refined data (Pfeffer: Introduction: “The algorithm is of hybrid quantum-classical nature since the optimization of the output map is done by a classical ridge regression.”), and repeating operations that produce classical output to gather statistics for post-processing refinement (Pfeffer: IV.C. “Each possible p-blocked reservoir configurations at n qubits was trained and then run for 100 different realizations to gather statistics.”).
As to Claim 20, it is rejected for similar reasons as claim 1.
B.
Claims 2-4, 10 and 19 are rejected under 35 U.S.C. § 103 as being unpatentable over non-patent literature Pfeffer et al. (“Pfeffer”), “Hybrid quantum-classical reservoir computing of thermal convection flow” published in 2022 in view of non-patent literature Govia et al. (“Govia”), “Nonlinear input transformations are ubiquitous in quantum reservoir computing,” published in 2022 in further of view non-patent literature Fry et al. (“Fry”), “Optimizing quantum noise‑induced reservoir computing for nonlinear and chaotic time series prediction,” published in 2023.
As to Claim 2, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer further teaches: forming a reservoir state vector based on one or more of the generated reservoir states, input data, and a bias term (Pfeffer: eq. 15 includes the generated reservoir state and the input data).
Pfeffer and Govia may not explicitly teach: wherein the method further comprises determining whether additional input data is to be processed.
Fry teaches in general concepts related to quantum noise-induced reservoir, in which reservoir noise is used as a resource to generate expressive, nonlinear signals that are efficiently learned with a single linear output layer (Fry: Abstract). Specifically, Fry teaches that optimization of the reservoir involves different stopping conditions, which would therefore determine whether additional input data is to be processed (Fry: “Noise optimization section,” ”Optimization algorithms require stopping conditions. The three stopping condition were: multiple small changes in MSE, long iteration runtime without update and the maximum number of iterations was 5, which is generally observed to be a large number for the optimization algorithms used”).
It would have been obvious to a person having ordinary skill in the art at a time before the effective filing date of the application to have modified the Pfeffer-Govia disclosures and teachings by allowing for stopping conditions in the iterations as taught by Fry. Such a person would have been motivated to do so with a reasonable expectation of success to allow for convergence of the reservoir to a final state for measurement.
As to Claim 3, Pfeffer, Govia and Fry teach the limitations of claim 2.
Pfeffer further teaches: wherein the method further comprises applying a ridge regression procedure to a series of the formed reservoir state vectors to determine a plurality of readout parameters (Pfeffer: Introduction section, “The algorithm is of hybrid quantum-classical nature since the optimization of the output map is done by a classical ridge regression.”; Sec. III.C., the readout requires K projective measurements and other parameters).
As to Claim 4, Pfeffer, Govia and Fry teach the limitations of claim 3.
Pfeffer further teaches: wherein the method further comprises generating predictions of a future state of the dynamical system based on the readout parameters and the reservoir state vector (Pfeffer: Sec III.C, the predictions are completed based on the readout hyperparameters, cost functions, etc) .
As to Claim 10, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer and Govia may not explicitly teach: wherein encoding the transformed data and encoding the measurement feedback includes a use of feature map functions for transforming data encoded parameters corresponding to a first set of parametrized layers and a second set of parameterized layers.
Govia does however further teach the use of a “of dynamical map of the input data into reservoir state variables, which we refer to as the input encoding. It has been shown for classical reservoirs that a nonlinear input encoding may be sufficient for task performance” (Govia: Sec. 1, Introduction). This dynamical map is encoding the transformed data (and the measurement feedback subsequently) by use of a mapping (Govia’s dynamical map).
Fry teaches in general concepts related to quantum noise-induced reservoir, in which reservoir noise is used as a resource to generate expressive, nonlinear signals that are efficiently learned with a single linear output layer (Fry: Abstract). Fry teaches the use of parameterized feature-maps implemented by parameterized RX gates (Fry: text around eq. 6, “entanglement scheme of RZZi,j gates, which are 2-qubit entangling gates where all RX(θ) and RZ(θ) rotation gates encode the time series data with a scaling map, θ = φ(u)”).
It would have been obvious to a person having ordinary skill in the art at a time before the effective filing date of the application to have modified the Pfeffer-Govia disclosures and teachings by utilizing feature maps in a parameterized fashion for each layer as taught and suggested by Fry. Such a person would have been motivated to do so with a reasonable expectation of success to allow for implementing a quantum noise-induced reservoir computing scheme.
As to Claim 19, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer and Govia may not explicitly teach: further comprising the steps of repeating quantum circuit operations with each repetition compiled differently to enable post-processing suppression techniques and augmenting a quantum output with a classical co-processor for system enhancement based on accumulated statistics.
Fry teaches in general concepts related to quantum noise-induced reservoir, in which reservoir noise is used as a resource to generate expressive, nonlinear signals that are efficiently learned with a single linear output layer (Fry: Abstract). Specifically, Fry teaches that optimization of the reservoir involves different stopping conditions, which would therefore determine whether additional input data is to be processed (Fry: “Noise optimization section,” ”Optimization algorithms require stopping conditions. The three stopping condition were: multiple small changes in MSE, long iteration runtime without update and the maximum number of iterations was 5, which is generally observed to be a large number for the optimization algorithms used”). Examiner asserts the observed number is based on accumulated statistic. Examiner also notes that the “to enable post-processing suppression technique” is functional language of limited patentable weight.
It would have been obvious to a person having ordinary skill in the art at a time before the effective filing date of the application to have modified the Pfeffer-Govia disclosures and teachings by allowing for repeating quantum circuit operations in the iterations as taught by Fry. Such a person would have been motivated to do so with a reasonable expectation of success to allow for convergence of the reservoir to a final state for measurement.
C.
Claim 8 is rejected under 35 U.S.C. § 103 as being unpatentable over non-patent literature Pfeffer et al. (“Pfeffer”), “Hybrid quantum-classical reservoir computing of thermal convection flow” published in 2022 in view of non-patent literature Govia et al. (“Govia”), “Nonlinear input transformations are ubiquitous in quantum reservoir computing,” published in 2022 in further of view non-patent literature Bogris et al. (“Bogris”), “Fabry-Perot Lasers as Enablers for Parallel Reservoir Computing,” published in 2021.
As to Claim 8, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer and Govia may not explicitly teach: wherein each of the plurality of transformation matrices comprise a random weight matrix or a Fourier-like matrix; and
the set of transformed data are generated by performing functional transformations over the first element of the input data.
Bogris teaches in general concepts related to the use of Fabry-Perot (FP) lasers as potential neuromorphic computing machines with parallel processing capabilities (Bogris: Abstract). Specifically Bogris teaches that a random weight matrix from the fixed input mask (i.e. transformation matrix) is used (Bogris: II.B, “The mask signal m(t) is the random connectivity matrix at the input with Nv values that take values from 0 to 1.”).
It would have been obvious to a person having ordinary skill in the art at a time before the effective filing date of the application to have modified the Pfeffer-Govia disclosures and teachings by utilizing a random weight matrix as taught and suggested by Bogris. Such a person would have been motivated to do so with a reasonable expectation of success to allow for implementing an optimized quantum reservoir computing scheme.
D.
Claims 16 is rejected under 35 U.S.C. § 103 as being unpatentable over non-patent literature Pfeffer et al. (“Pfeffer”), “Hybrid quantum-classical reservoir computing of thermal convection flow” published in 2022 in view of non-patent literature Govia et al. (“Govia”), “Nonlinear input transformations are ubiquitous in quantum reservoir computing,” published in 2022 in further of view non-patent literature Strock et al. (“Strock”), “A Simple Reservoir Model of Working Memory with Real Values” published in 2018.
As to Claim 16, Pfeffer and Govia teach the limitations of claim 1.
Pfeffer and Govia may not explicitly teach: wherein each of the transformation matrices is a fixed transformation matrix associated with a reservoir state, a measured state, and an input state.
Strock teaches in general concepts related to an echo state network (ESN) that is reservoir computing (Strock: Abstract). Specifically, Strock teaches that fixed weights (i.e. represented in matrices) for input, recurrent and feedback are used (Strock: Sec. II.A., “only the output weights will be learned; input, recurrent and feedback weight are generated randomly and kept fixed.”). Matrices to model the ESN are modeled with respective matrices for the input, feedback and output weight matrices (Strock: Sec. II.A., eqs. 1-3).
It would have been obvious to a person having ordinary skill in the art at a time before the effective filing date of the application to have modified the Pfeffer-Govia disclosures and teachings by utilizing fixed weight matrices for the reservoir state, measured state and input states as taught and suggested by Strock. Such a person would have been motivated to do so with a reasonable expectation of success to allow for a modeling of the properties of working memory mechanisms (Strock: Sec. II).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAMES T TSAI whose telephone number is (571)270-3916. The examiner can normally be reached M-F 8-5 Eastern.
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, Viker Lamardo can be reached at 571-270-5871. 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.
/JAMES T TSAI/ Primary Examiner, Art Unit 2147
Prosecution Insights