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 .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 11/30/2023 and 04/30/2026 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Objections
Claims 13-20 are objected to because of the following informalities:
In claim 13, line 2, “a TQFM” recites an acronym which has not been spelled out at least once for the first citation in the claim. The claim should be amended to read “a temporal quantum feature map (TQFM)”
In claim 19, line 1, “a TQFM” recites an acronym which has not been spelled out at least once for the first citation in the claim. The claim should be amended to read “a temporal quantum feature map (TQFM)”
Dependent claims 14-18 are objected based on being directly or indirectly dependent on objected claim 13.
Dependent claim 20 is objected based on being directly or indirectly dependent on objected claim 19.
Appropriate correction is required.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 3-5, 9-13, 15-17, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Bocharov et al. (US 2021/0089953 A1) in view of Markov et al. ("Implementation and Learning of Quantum Hidden Markov Models").
Regarding Claim 1,
Bocharov et al. teaches a system, comprising: a memory that stores computer-executable components; and a processor that executes the computer-executable components stored in the memory (Fig. 6; [0063]: "FIG. 6 illustrates example operations 600 for training and constructing a quantum prediction circuit capable of predicting a future data bit in a time-sequential sequence" teaches a method for sequential data prediction using a quantum circuit. [0041]: "Given an encoded feature vector φ(x)=|b0, b1, b2, b3, b4, b5, b6, b7>, (the “input qubit register”), the gates (G) of the quantum prediction circuit 400 execute a sequence of quantum state transformations that map the encoded feature vector φ(x) to an output vector φ′(x)=Uθφ(x) by applying a unitary operation Uθ, which is parameterized by a set of variables θj. In the example quantum prediction circuit 400, the input feature vector φ(x) is a quantum state vector encoding a sequential subset of a time-sequential sequence" teaches that the time-sequential input sequence is encoded into a feature vector (e.g. using temporal quantum feature map) in the quantum circuit. Fig. 9; [0079]-[0080]: "With reference to FIG. 9, an exemplary system for implementing the disclosed technology includes a general purpose computing device in the form of an exemplary conventional PC 900, including one or more processing units 902, a system memory 904, and a system bus 906 that couples various system components including the system memory 904 to the one or more processing units 902 … the system memory 904 stores gate parameter definitions and hyperparameters that are used to configure a quantum computer to predict future bits in a time-sequential series of data points. Computer-executable instructions are also stored for receiving precisions as well as communicating circuit definitions and states to be used" teaches that the method may be implemented by a system comprising processing units (processor) and memory storing computer-executable instructions (computer-executable components)).
Bocharov et al. does not appear to explicitly teach a computation component that uses a temporal quantum feature map (TQFM) to compute a kernel element between two sequences of symbols by respectively processing two input sequences as vectors.
However, Markov et al. teaches a computation component that uses a temporal quantum feature map (TQFM) to compute a kernel element between two sequences of symbols by respectively processing two input sequences as vectors (Section II. A. 3: "The POVM operations described in Section II A 2 define a joint stochastic process of quantum state transitions and symbols emission. The quantum operations’ action on a quantum state defining the process resembles the classic observable operators’ action on a classic stochastic state as defined in Section II A 1. The similarities lead to the following quantum HMM definition: Definition 2 (Quantum Hidden Markov Model). A quantum HMM (QHMM) Q over a finite alphabet of observable symbols Σ and finite N-dimensional Hilbert space is a 4-tuple: Q={Σ,HS,T ={Ta}a∈Σ,ρ0} where Σ is a finite alphabet of observable symbols, HS is a finite dimensional Hilbert space defining the model’s states as density operators, T is a set of non-trace-increasing quantum operators Ta: D(HS) → D(HS), indexed by the symbols of the alphabet Σ, and ρ0 is an initial density operator. When the operation T = {Ta}a∈Σ is applied to the current state ρ, a symbol a will be observed/emitted … Applying the operation repeatedly t times will emit a sequence of symbols a = a1...at with probability p[a] =Tr(Tat ...Ta1 ρ0). EQ 14 above shows that QHMM Q defines a sequence function fQ … For any integer t the sequence function (EQ 15) defines the distribution DtQ over the sequences of length t … Therefore, a QHMM Q defines a stochastic process language LQ over the set of finite sequences Σ*" teaches a QHMM (TQFM) that computes a quantum operation (e.g. computes a kernel element) between sequences of symbols by processing input sequences. Section II. A. 1, first paragraph: "We study a class of discrete time stationary stochastic processes, {Yt|t ∈ N,Yt ∈ Σ}, where Σ = {a1...am} is a finite set of observable symbols, called an alphabet. The set of all finite sequences over the alphabet Σ, including the empty sequence ϵ, is denoted by Σ*. For any a ∈ Σ* let |a| be the number of symbols in the sequence. If p and s are sequences, then ps denotes their concatenation. The sequence p is called the prefix and s is called the suffix of ps. Any subset of Σ* is a language L over the alphabet. The sequences belonging to a language are referred to as words. The set of sequences originated by observations or measurement of the evolution of a discrete time process is called the process language" teaches that the input sequences includes two sequences of symbols. Section III. F, second paragraph: "The second method we employ in our analysis is the use of an ansatz (or parametric quantum circuit) template, built from variational gates, to approximate an QHMM. The topic of efficient ansatz selection is an area of active research. Due to the presence of noise in the near term quantum devices, one must select ansatz which is sufficiently expressive (i.e. able to access the solution space in the Hilbert space) while maintaining low parameter count and lower depth of the quantum circuits in order to suppress the noise. Moreover, the proposed ansatz should contain sufficient entanglement to be non-simulable on a classical computer ... Recently, another circuit (FIG 14) was used for kernel alignment" teaches that the QHMM (TQFM) is implemented using an ansatz quantum circuit and used for kernel alignment (e.g. QHMM is used for kernel computation)).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate a computation component that uses a temporal quantum feature map (TQFM) to compute a kernel element between two sequences of symbols by respectively processing two input sequences as vectors as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 3,
Bocharov et al. in view of Markov et al. teaches the system of claim 1.
In addition, Markov et al. further teaches wherein the two input sequences are processed as respective vectors having equal lengths (Section II. A. 1, first-second paragraphs: "The set of all finite sequences over the alphabet Σ, including the empty sequence ϵ, is denoted by Σ*. For any a ∈ Σ* let |a| be the number of symbols in the sequence. If p and s are sequences, then ps denotes their concatenation. The sequence p is called the prefix and s is called the suffix of ps. Any subset of Σ* is a language L over the alphabet. The sequences belonging to a language are referred to as words. The set of sequences originated by observations or measurement of the evolution of a discrete time process is called the process language ... A stochastic process language L is a process language together with a set of probability distributions DtL, each of which is defined on sequences with length t. The following properties of a stochastic process language are straightforward to prove: ... 2. Every stochastic language L and sequence of symbols p ∈ Σ* define collections of distributions over the sequences with equal length:
PNG
media_image1.png
58
316
media_image1.png
Greyscale
We call two sequences equivalent w.r.t a stochastic language L if they define the same distribution over Σt, ∀t ∈ N" teaches that the prefixes and suffixes sequences (two input sequences) can have equal lengths).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the two input sequences are processed as respective vectors having equal lengths as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 4,
Bocharov et al. in view of Markov et al. teaches the system of claim 1.
In addition, Markov et al. further teaches wherein the two input sequences are processed as respective vectors having different lengths (Table III; Section II. C, second to last paragraph: "When working with sequences, it is often convenient to present probabilities in the form of the Hankel matrix that separates the process into prefixes and suffixes of varying lengths. The Hankel matrix for the current process is shown in Table III" teaches that the prefixes and suffixes sequences (two input sequences) can have varying (different) lengths).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the two input sequences are processed as respective vectors having different lengths as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 5,
Bocharov et al. in view of Markov et al. teaches the system of claim 1.
In addition, Markov et al. further teaches wherein the TQFM is a quantum channel (Section II. A. 4, first paragraph: "The definition of a QHMM in the previous section is based on the concept of a POVM operation on a quantum state. The observable symbols generation process is encoded in the operational elements (Kraus operators) of the quantum operation. This approach is very convenient for understanding QHMMs as quantum information processing channels, for studying of their informational complexity, expressive capacity, and connections to the stochastic process languages and corresponding automata" teaches that the QHMM (TFQM) is a quantum information processing channel (quantum channel)).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the TQFM is a quantum channel as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 9,
Bocharov et al. in view of Markov et al. teaches the system of claim 5.
In addition, Markov et al. further teaches further comprising: an optimization component that parametrizes the quantum channel to optimize the quantum channel using an optimization algorithm and generate an optimized quantum channel (Algorithm 2; Section III. E: "In this paper we propose the following algorithm for learning quantum hidden Markov model using ansatz circuits … 3 Find optimal parameters for quantum circuit with classical technique, e.g. Nelder–Mead method" teaches that parameters of the quantum hidden Markov model (quantum channel) are optimized using the Nelder–Mead method (optimization algorithm) to generate an optimized quantum circuit for the quantum hidden Markov model (optimized quantum channel)).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising: an optimization component that parametrizes the quantum channel to optimize the quantum channel using an optimization algorithm and generate an optimized quantum channel. as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 10,
Bocharov et al. in view of Markov et al. teaches the system of claim 9.
In addition, Markov et al. further teaches further comprising: a training component that trains a model for sequence prediction using the optimized quantum channel (Algorithm 2; Section III. E: "In this paper we propose the following algorithm for learning quantum hidden Markov model using ansatz circuits … 3 Find optimal parameters for quantum circuit with classical technique, e.g. Nelder–Mead method ... Output: trained circuit that can be extended to the required sequence length" teaches learning (training) the model using the optimized parameters of the quantum circuit (optimized quantum channel) to obtain a trained circuit for implementing the model. Fig. 16; Section IV. A, fifth paragraph: "The final circuit for 2 step sequence is shown on FIG 15. Comparison of observed and predicted sequence probabilities is shown on FIG 16" teaches that the final (trained) circuit is used to predict sequence probabilities (sequence prediction)).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising: a training component that trains a model for sequence prediction using the optimized quantum channel as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 11,
Bocharov et al. in view of Markov et al. teaches the system of claim 1.
In addition, Markov et al. further teaches wherein a number of qubits of the TQFM is a hyperparameter that is not directly determined by a length of an input sequence to be embedded (Section III. B, first paragraph: "The hypotheses space of the learning problem contains the QHMMs defined by the unitary definition (Definition 3): Q={Σ,HS,HE,U,M,ρ0} where the parameters are specified and restricted according to the observed stochastic process sequences: • The alphabet Σ contains the observed m symbols. • HS is a Hilbert space with dimension
N
^
= ⌈ (
r
^
)⌉ where
r
^
is the estimated maximal rank of the Hankel matrix built for the data sample Y. • HE is a Hilbert space with dimension M = ⌈log2m⌉. We select any orthonormal basis E ={|e0⟩···|eM−1⟩} of HE. • U is a unitary operation on the Hilbert space HS ⊗ HE implemented by a quantum circuit of log2
N
^
+log2M qubits" teaches that the number of qubits of the QHMM (TFQM) is not directly determined by a length of an input sequence to be embedded).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein a number of qubits of the TQFM is a hyperparameter that is not directly determined by a length of an input sequence to be embedded as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 12,
Bocharov et al. in view of Markov et al. teaches the system of claim 1.
In addition, Markov et al. further teaches wherein a longer input sequence is included with subsequent input sequences for training the TQFM (Algorithm 2; Section III. E: "In this paper we propose the following algorithm for learning quantum hidden Markov model using ansatz circuits … Input: table of observed sequences and corresponding probabilities, min-max sequence length ... Output: trained circuit that can be extended to the required sequence length" teaches learning (training) the QHMM (TQFM) using the observed (input) sequences from min to max sequence length (e.g. a longer/max length sequence is included in the subsequent input sequences for training).), and
wherein a resulting quantum state is a pure quantum state or a mixed quantum state (Section II. A. 2, first paragraph: "We model the state of a stochastic process with an n-qubit quantum system and the associated complex Hilbert space HS. The vectors of HS will be denoted by |v⟩. A complete description of the quantum system at a moment in time called a state. We will consider a space of states which is a convex subspace of HS. The extreme points and mixed points of this subspace are called pure and mixed states respectively" teaches that the quantum states are either pure quantum states or mixed quantum states).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein a longer input sequence is included with subsequent input sequences for training the TQFM, and wherein a resulting quantum state is a pure quantum state or a mixed quantum state as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 13,
Bocharov et al. teaches a computer-implemented method, comprising: … a system operatively coupled to a processor (Fig. 6; [0063]: "FIG. 6 illustrates example operations 600 for training and constructing a quantum prediction circuit capable of predicting a future data bit in a time-sequential sequence" teaches a method for sequential data prediction using a quantum circuit. [0041]: "Given an encoded feature vector φ(x)=|b0, b1, b2, b3, b4, b5, b6, b7>, (the “input qubit register”), the gates (G) of the quantum prediction circuit 400 execute a sequence of quantum state transformations that map the encoded feature vector φ(x) to an output vector φ′(x)=Uθφ(x) by applying a unitary operation Uθ, which is parameterized by a set of variables θj. In the example quantum prediction circuit 400, the input feature vector φ(x) is a quantum state vector encoding a sequential subset of a time-sequential sequence" teaches that the time-sequential input sequence is encoded into a feature vector (e.g. using temporal quantum feature map) in the quantum circuit. Fig. 9; [0079]: "With reference to FIG. 9, an exemplary system for implementing the disclosed technology includes a general purpose computing device in the form of an exemplary conventional PC 900, including one or more processing units 902, a system memory 904, and a system bus 906 that couples various system components including the system memory 904 to the one or more processing units 902" teaches that the method may be implemented by a computing device (e.g. computer-implemented) with a processing unit (processor)).
Bocharov et al. does not appear to explicitly teach using, …, a TQFM to compute a kernel element between two sequences of symbols by respectively processing two input sequences as vectors.
However, Markov et al. teaches using, …, a TQFM to compute a kernel element between two sequences of symbols by respectively processing two input sequences as vectors (Section II. A. 3: "The POVM operations described in Section II A 2 define a joint stochastic process of quantum state transitions and symbols emission. The quantum operations’ action on a quantum state defining the process resembles the classic observable operators’ action on a classic stochastic state as defined in Section II A 1. The similarities lead to the following quantum HMM definition: Definition 2 (Quantum Hidden Markov Model). A quantum HMM (QHMM) Q over a finite alphabet of observable symbols Σ and finite N-dimensional Hilbert space is a 4-tuple: Q={Σ,HS,T ={Ta}a∈Σ,ρ0} where Σ is a finite alphabet of observable symbols, HS is a finite dimensional Hilbert space defining the model’s states as density operators, T is a set of non-trace-increasing quantum operators Ta: D(HS) → D(HS), indexed by the symbols of the alphabet Σ, and ρ0 is an initial density operator. When the operation T = {Ta}a∈Σ is applied to the current state ρ, a symbol a will be observed/emitted … Applying the operation repeatedly t times will emit a sequence of symbols a = a1...at with probability p[a] =Tr(Tat ...Ta1 ρ0). EQ 14 above shows that QHMM Q defines a sequence function fQ … For any integer t the sequence function (EQ 15) defines the distribution DtQ over the sequences of length t … Therefore, a QHMM Q defines a stochastic process language LQ over the set of finite sequences Σ*" teaches a QHMM (TQFM) that computes a quantum operation (e.g. computes a kernel element) between sequences of symbols by processing input sequences. Section II. A. 1, first paragraph: "We study a class of discrete time stationary stochastic processes, {Yt|t ∈ N,Yt ∈ Σ}, where Σ = {a1...am} is a finite set of observable symbols, called an alphabet. The set of all finite sequences over the alphabet Σ, including the empty sequence ϵ, is denoted by Σ*. For any a ∈ Σ* let |a| be the number of symbols in the sequence. If p and s are sequences, then ps denotes their concatenation. The sequence p is called the prefix and s is called the suffix of ps. Any subset of Σ* is a language L over the alphabet. The sequences belonging to a language are referred to as words. The set of sequences originated by observations or measurement of the evolution of a discrete time process is called the process language" teaches that the input sequences includes two sequences of symbols. Section III. F, second paragraph: "The second method we employ in our analysis is the use of an ansatz (or parametric quantum circuit) template, built from variational gates, to approximate an QHMM. The topic of efficient ansatz selection is an area of active research. Due to the presence of noise in the near term quantum devices, one must select ansatz which is sufficiently expressive (i.e. able to access the solution space in the Hilbert space) while maintaining low parameter count and lower depth of the quantum circuits in order to suppress the noise. Moreover, the proposed ansatz should contain sufficient entanglement to be non-simulable on a classical computer ... Recently, another circuit (FIG 14) was used for kernel alignment" teaches that the QHMM (TQFM) is implemented using an ansatz quantum circuit and used for kernel alignment (e.g. QHMM is used for kernel computation)).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate using, …, a TQFM to compute a kernel element between two sequences of symbols by respectively processing two input sequences as vectors as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 15,
Bocharov et al. in view of Markov et al. teaches the computer-implemented method of claim 13.
In addition, Markov et al. further teaches wherein the two input sequences are processed as respective vectors having equal lengths (Section II. A. 1, first-second paragraphs: "The set of all finite sequences over the alphabet Σ, including the empty sequence ϵ, is denoted by Σ*. For any a ∈ Σ* let |a| be the number of symbols in the sequence. If p and s are sequences, then ps denotes their concatenation. The sequence p is called the prefix and s is called the suffix of ps. Any subset of Σ* is a language L over the alphabet. The sequences belonging to a language are referred to as words. The set of sequences originated by observations or measurement of the evolution of a discrete time process is called the process language ... A stochastic process language L is a process language together with a set of probability distributions DtL, each of which is defined on sequences with length t. The following properties of a stochastic process language are straightforward to prove: ... 2. Every stochastic language L and sequence of symbols p ∈ Σ* define collections of distributions over the sequences with equal length:
PNG
media_image1.png
58
316
media_image1.png
Greyscale
We call two sequences equivalent w.r.t a stochastic language L if they define the same distribution over Σt, ∀t ∈ N" teaches that the prefixes and suffixes sequences (two input sequences) can have equal lengths).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the two input sequences are processed as respective vectors having equal lengths as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 16,
Bocharov et al. in view of Markov et al. teaches the computer-implemented method of claim 13.
In addition, Markov et al. further teaches wherein the two input sequences are processed as respective vectors having different lengths (Table III; Section II. C, second to last paragraph: "When working with sequences, it is often convenient to present probabilities in the form of the Hankel matrix that separates the process into prefixes and suffixes of varying lengths. The Hankel matrix for the current process is shown in Table III" teaches that the prefixes and suffixes sequences (two input sequences) can have varying (different) lengths).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the two input sequences are processed as respective vectors having different lengths as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 17,
Bocharov et al. in view of Markov et al. teaches the computer-implemented method of claim 13.
In addition, Markov et al. further teaches wherein the TQFM is a quantum channel (Section II. A. 4, first paragraph: "The definition of a QHMM in the previous section is based on the concept of a POVM operation on a quantum state. The observable symbols generation process is encoded in the operational elements (Kraus operators) of the quantum operation. This approach is very convenient for understanding QHMMs as quantum information processing channels, for studying of their informational complexity, expressive capacity, and connections to the stochastic process languages and corresponding automata" teaches that the QHMM (TFQM) is a quantum information processing channel (quantum channel)).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the TQFM is a quantum channel as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Regarding Claim 19,
Bocharov et al. teaches a computer program product for employing a TQFM for kernel-based sequential data prediction, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor (Fig. 6; [0063]: "FIG. 6 illustrates example operations 600 for training and constructing a quantum prediction circuit capable of predicting a future data bit in a time-sequential sequence" teaches a method for sequential data prediction using a quantum circuit. [0041]: "Given an encoded feature vector φ(x)=|b0, b1, b2, b3, b4, b5, b6, b7>, (the “input qubit register”), the gates (G) of the quantum prediction circuit 400 execute a sequence of quantum state transformations that map the encoded feature vector φ(x) to an output vector φ′(x)=Uθφ(x) by applying a unitary operation Uθ, which is parameterized by a set of variables θj. In the example quantum prediction circuit 400, the input feature vector φ(x) is a quantum state vector encoding a sequential subset of a time-sequential sequence" teaches that the time-sequential input sequence is encoded into a feature vector (e.g. using temporal quantum feature map) in the quantum circuit. Fig. 9; [0081]-[0082]: "The exemplary PC 900 further includes one or more storage devices 930 such as a hard disk drive for reading from and writing to a hard disk, a magnetic disk drive for reading from or writing to a removable magnetic disk, and an optical disk drive for reading from or writing to a removable optical disk (such as a CD-ROM or other optical media).-Such storage devices can be connected to the system bus 906 by a hard disk drive interface, a magnetic disk drive interface, and an optical drive interface, respectively. The drives and their associated computer readable media provide nonvolatile storage of computer-readable instructions, data structures, program modules, and other data for the PC 900 … A number of program modules may be stored in the storage devices 930 including an operating system, one or more application programs, other program modules, and program data. Storage of computer-executable instructions for training procedures and configuring a quantum computer can be stored in the storage devices 930 as well as or in addition to the memory 904" teaches that the method may be implemented by program modules (computer program product) comprising computer readable media (computer readable storage medium) storing computer-executable instructions (program instructions) for execution by a computer (processor)).
Bocharov et al. does not appear to explicitly teach use, by the processor, a TQFM to compute a kernel element between two sequences of symbols, on a quantum computer, by respectively processing two input sequences as vectors.
However, Markov et al. teaches use, by the processor, a TQFM to compute a kernel element between two sequences of symbols, on a quantum computer, by respectively processing two input sequences as vectors (Section II. A. 3: "The POVM operations described in Section II A 2 define a joint stochastic process of quantum state transitions and symbols emission. The quantum operations’ action on a quantum state defining the process resembles the classic observable operators’ action on a classic stochastic state as defined in Section II A 1. The similarities lead to the following quantum HMM definition: Definition 2 (Quantum Hidden Markov Model). A quantum HMM (QHMM) Q over a finite alphabet of observable symbols Σ and finite N-dimensional Hilbert space is a 4-tuple: Q={Σ,HS,T ={Ta}a∈Σ,ρ0} where Σ is a finite alphabet of observable symbols, HS is a finite dimensional Hilbert space defining the model’s states as density operators, T is a set of non-trace-increasing quantum operators Ta: D(HS) → D(HS), indexed by the symbols of the alphabet Σ, and ρ0 is an initial density operator. When the operation T = {Ta}a∈Σ is applied to the current state ρ, a symbol a will be observed/emitted … Applying the operation repeatedly t times will emit a sequence of symbols a = a1...at with probability p[a] =Tr(Tat ...Ta1 ρ0). EQ 14 above shows that QHMM Q defines a sequence function fQ … For any integer t the sequence function (EQ 15) defines the distribution DtQ over the sequences of length t … Therefore, a QHMM Q defines a stochastic process language LQ over the set of finite sequences Σ*" teaches a QHMM (TQFM) that computes a quantum operation (e.g. computes a kernel element) between sequences of symbols by processing input sequences. Section II. A. 1, first paragraph: "We study a class of discrete time stationary stochastic processes, {Yt|t ∈ N,Yt ∈ Σ}, where Σ = {a1...am} is a finite set of observable symbols, called an alphabet. The set of all finite sequences over the alphabet Σ, including the empty sequence ϵ, is denoted by Σ*. For any a ∈ Σ* let |a| be the number of symbols in the sequence. If p and s are sequences, then ps denotes their concatenation. The sequence p is called the prefix and s is called the suffix of ps. Any subset of Σ* is a language L over the alphabet. The sequences belonging to a language are referred to as words. The set of sequences originated by observations or measurement of the evolution of a discrete time process is called the process language" teaches that the input sequences includes two sequences of symbols. Section III. F, second paragraph: "The second method we employ in our analysis is the use of an ansatz (or parametric quantum circuit) template, built from variational gates, to approximate an QHMM. The topic of efficient ansatz selection is an area of active research. Due to the presence of noise in the near term quantum devices, one must select ansatz which is sufficiently expressive (i.e. able to access the solution space in the Hilbert space) while maintaining low parameter count and lower depth of the quantum circuits in order to suppress the noise. Moreover, the proposed ansatz should contain sufficient entanglement to be non-simulable on a classical computer ... Recently, another circuit (FIG 14) was used for kernel alignment" teaches that the QHMM (TQFM) is implemented using an ansatz quantum circuit and used for kernel alignment (e.g. QHMM is used for kernel computation)).
Bocharov et al. and Markov et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate use, by the processor, a TQFM to compute a kernel element between two sequences of symbols, on a quantum computer, by respectively processing two input sequences as vectors as taught by Markov et al. to the disclosed invention of Bocharov et al.
One of ordinary skill in the art would have been motivated to make this modification to provide "the advantages of using quantum hidden Markov models over classical counterparts … [and provide] a practical, hardware efficient quantum circuit ansatz, as well as a training algorithm" (Markov et al. Abstract).
Claims 2, 6-7, 14, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Bocharov et al. (US 2021/0089953 A1) in view of Markov et al. ("Implementation and Learning of Quantum Hidden Markov Models") and further in view of Monras et al. ("Hidden Quantum Markov Models and non-adaptive read-out of many-body states").
Regarding Claim 2,
Bocharov et al. in view of Markov et al. teaches the system of claim 1.
Bocharov et al. in view of Markov et al. does not appear to explicitly teach wherein the two input sequences respectively represent two input time series, and wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols.
However, Monras et al. teaches wherein the two input sequences respectively represent two input time series (Section 2, first-third paragraphs: "Consider the temporal evolution of some natural system. The evolution is monitored by a series of measurements—numbers sequentially registered. Considering the numbers to be discrete, each such measurement can be taken as a discrete random variable Xi. The probability distribution Pr(
X
-
∞
+
∞
) over bi-infinite sequences
X
-
∞
+
∞
of these random variables is what we refer to as a stochastic process. Such a process is a complete description of a system’s behavior … Among all possible stochastic processes, several models have proven successful in order to model a wide range of practical situations. One such class of processes are the so-called Hidden Markov Models (HMM from now on). There are several possible ways to define an HMM. The most common among mathematicians is as a stochastic process S for which every symbol sn is generated conditionally from a Markovian process X, with probability Pr(sn|xn)" teaches that the input sequences represent temporal series measurements (input time series). Section 3, first paragraph: "We now turn to quantum systems and the process languages they generate. As with stochastic processes, we assume that the evolution of a quantum system is recorded as a sequence of measurement outcomes. Just like before, the distribution over sequences of these random variables is a process" teaches that the sequences for the quantum system are sequences of measurement outcomes (time series)), and
wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols (Section 2, first-third paragraphs: "Consider the temporal evolution of some natural system. The evolution is monitored by a series of measurements—numbers sequentially registered. Considering the numbers to be discrete, each such measurement can be taken as a discrete random variable Xi. The probability distribution Pr(
X
-
∞
+
∞
) over bi-infinite sequences
X
-
∞
+
∞
of these random variables is what we refer to as a stochastic process. Such a process is a complete description of a system’s behavior … Among all possible stochastic processes, several models have proven successful in order to model a wide range of practical situations. One such class of processes are the so-called Hidden Markov Models (HMM from now on). There are several possible ways to define an HMM. The most common among mathematicians is as a stochastic process S for which every symbol sn is generated conditionally from a Markovian process X, with probability Pr(sn|xn)" teaches that the input sequences represent temporal series measurements (input time series). Section 3, first paragraph: "We now turn to quantum systems and the process languages they generate. As with stochastic processes, we assume that the evolution of a quantum system is recorded as a sequence of measurement outcomes. Just like before, the distribution over sequences of these random variables is a process" teaches that the sequences for the quantum system are sequences of measurement outcomes (time series). Section 4.2, second paragraph: "A Hidden Quantum Markov Model (HQMM) is a d-level quantum system ρ together with a set of quantum operations Ks such that ΣsKs is trace-preserving. At every time step a symbol is generated with probability Pr(s) = tr[Ksρ] and the state vector is updated to ρs = Ksρ/Pr(s)" teaches that the HQMM quantum operation (e.g. computing kernel element) comprise generating a symbol at every time step (e.g. input time series is split into sequence of symbols)).
Bocharov et al., Markov et al., and Monras et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the two input sequences respectively represent two input time series, and wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols as taught by Monras et al. to the disclosed invention of Bocharov et al. in view of Markov et al.
One of ordinary skill in the art would have been motivated to make this modification because "the advantages of considering HQMMs are many-fold. On the one hand, it is sufficiently general to naturally encompass all Hidden Markov Models, a highly desirable feature for any quantum model attempting to outperform their classical counterpart. On the other hand, they provide the most compressed description of a quantum process by only accounting for the dynamics of the relevant part of the system, i.e., that which contains the quantum information latent in the system" (Monras et al. Section 7, first paragraph).
Regarding Claim 6,
Bocharov et al. in view of Markov et al. teaches the system of claim 5.
Bocharov et al. in view of Markov et al. does not appear to explicitly teach wherein the quantum channel is completely positive (CP).
However, Monras et al. teaches wherein the quantum channel is completely positive (CP) (Table 1; Section 4.2, first-second paragraphs: "Notice in Table 1 the analogy between the structure of stochastic generators and the notion of stochastic quantum operation. Every piece in the classical formalism has a direct counterpart in the quantum formalism. It is thus natural to define a quantum stochastic generator following this analogy ... Definition: A Hidden Quantum Markov Model (HQMM) is a d-level quantum system ρ together with a set of quantum operations Ks such that ∑sKs is trace-preserving ... Analogy between the classical finite-state generator and the stochastic quantum operation. The mapping is straightforward by replacing probability vectors by density operators, transition matrices with CP-maps, of which stochastic matrices correspond to trace-preserving maps" teaches that the HQMM (quantum channel) is completely positive (CP)).
Bocharov et al., Markov et al., and Monras et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the quantum channel is completely positive (CP) as taught by Monras et al. to the disclosed invention of Bocharov et al. in view of Markov et al.
One of ordinary skill in the art would have been motivated to make this modification because "the advantages of considering HQMMs are many-fold. On the one hand, it is sufficiently general to naturally encompass all Hidden Markov Models, a highly desirable feature for any quantum model attempting to outperform their classical counterpart. On the other hand, they provide the most compressed description of a quantum process by only accounting for the dynamics of the relevant part of the system, i.e., that which contains the quantum information latent in the system" (Monras et al. Section 7, first paragraph).
Regarding Claim 7,
Bocharov et al. in view of Markov et al. teaches the system of claim 5.
Bocharov et al. in view of Markov et al. does not appear to explicitly teach wherein the quantum channel is completely positive and trace preserving (CPTP).
However, Monras et al. teaches wherein the quantum channel is completely positive and trace preserving (CPTP) (Table 1; Section 4.2, first-second paragraphs: "Notice in Table 1 the analogy between the structure of stochastic generators and the notion of stochastic quantum operation. Every piece in the classical formalism has a direct counterpart in the quantum formalism. It is thus natural to define a quantum stochastic generator following this analogy ... Definition: A Hidden Quantum Markov Model (HQMM) is a d-level quantum system ρ together with a set of quantum operations Ks such that ∑sKs is trace-preserving ... Analogy between the classical finite-state generator and the stochastic quantum operation. The mapping is straightforward by replacing probability vectors by density operators, transition matrices with CP-maps, of which stochastic matrices correspond to trace-preserving maps" teaches that the HQMM (quantum channel) is completely positive and trace preserving (CPTP)).
Bocharov et al., Markov et al., and Monras et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the quantum channel is completely positive and trace preserving (CPTP) as taught by Monras et al. to the disclosed invention of Bocharov et al. in view of Markov et al.
One of ordinary skill in the art would have been motivated to make this modification because "the advantages of considering HQMMs are many-fold. On the one hand, it is sufficiently general to naturally encompass all Hidden Markov Models, a highly desirable feature for any quantum model attempting to outperform their classical counterpart. On the other hand, they provide the most compressed description of a quantum process by only accounting for the dynamics of the relevant part of the system, i.e., that which contains the quantum information latent in the system" (Monras et al. Section 7, first paragraph).
Regarding Claim 14,
Bocharov et al. in view of Markov et al. teaches the computer-implemented method of claim 13.
Bocharov et al. in view of Markov et al. does not appear to explicitly teach wherein the two input sequences respectively represent two input time series, and wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols.
However, Monras et al. teaches wherein the two input sequences respectively represent two input time series (Section 2, first-third paragraphs: "Consider the temporal evolution of some natural system. The evolution is monitored by a series of measurements—numbers sequentially registered. Considering the numbers to be discrete, each such measurement can be taken as a discrete random variable Xi. The probability distribution Pr(
X
-
∞
+
∞
) over bi-infinite sequences
X
-
∞
+
∞
of these random variables is what we refer to as a stochastic process. Such a process is a complete description of a system’s behavior … Among all possible stochastic processes, several models have proven successful in order to model a wide range of practical situations. One such class of processes are the so-called Hidden Markov Models (HMM from now on). There are several possible ways to define an HMM. The most common among mathematicians is as a stochastic process S for which every symbol sn is generated conditionally from a Markovian process X, with probability Pr(sn|xn)" teaches that the input sequences represent temporal series measurements (input time series). Section 3, first paragraph: "We now turn to quantum systems and the process languages they generate. As with stochastic processes, we assume that the evolution of a quantum system is recorded as a sequence of measurement outcomes. Just like before, the distribution over sequences of these random variables is a process" teaches that the sequences for the quantum system are sequences of measurement outcomes (time series)), and
wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols (Section 2, first-third paragraphs: "Consider the temporal evolution of some natural system. The evolution is monitored by a series of measurements—numbers sequentially registered. Considering the numbers to be discrete, each such measurement can be taken as a discrete random variable Xi. The probability distribution Pr(
X
-
∞
+
∞
) over bi-infinite sequences
X
-
∞
+
∞
of these random variables is what we refer to as a stochastic process. Such a process is a complete description of a system’s behavior … Among all possible stochastic processes, several models have proven successful in order to model a wide range of practical situations. One such class of processes are the so-called Hidden Markov Models (HMM from now on). There are several possible ways to define an HMM. The most common among mathematicians is as a stochastic process S for which every symbol sn is generated conditionally from a Markovian process X, with probability Pr(sn|xn)" teaches that the input sequences represent temporal series measurements (input time series). Section 3, first paragraph: "We now turn to quantum systems and the process languages they generate. As with stochastic processes, we assume that the evolution of a quantum system is recorded as a sequence of measurement outcomes. Just like before, the distribution over sequences of these random variables is a process" teaches that the sequences for the quantum system are sequences of measurement outcomes (time series). Section 4.2, second paragraph: "A Hidden Quantum Markov Model (HQMM) is a d-level quantum system ρ together with a set of quantum operations Ks such that ΣsKs is trace-preserving. At every time step a symbol is generated with probability Pr(s) = tr[Ksρ] and the state vector is updated to ρs = Ksρ/Pr(s)" teaches that the HQMM quantum operation (e.g. computing kernel element) comprise generating a symbol at every time step (e.g. input time series is split into sequence of symbols)).
Bocharov et al., Markov et al., and Monras et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the two input sequences respectively represent two input time series, and wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols as taught by Monras et al. to the disclosed invention of Bocharov et al. in view of Markov et al.
One of ordinary skill in the art would have been motivated to make this modification because "the advantages of considering HQMMs are many-fold. On the one hand, it is sufficiently general to naturally encompass all Hidden Markov Models, a highly desirable feature for any quantum model attempting to outperform their classical counterpart. On the other hand, they provide the most compressed description of a quantum process by only accounting for the dynamics of the relevant part of the system, i.e., that which contains the quantum information latent in the system" (Monras et al. Section 7, first paragraph).
Regarding Claim 20,
Bocharov et al. in view of Markov et al. teaches the computer program product of claim 19.
Bocharov et al. in view of Markov et al. does not appear to explicitly teach wherein the two input sequences respectively represent two input time series, and wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols.
However, Monras et al. teaches wherein the two input sequences respectively represent two input time series (Section 2, first-third paragraphs: "Consider the temporal evolution of some natural system. The evolution is monitored by a series of measurements—numbers sequentially registered. Considering the numbers to be discrete, each such measurement can be taken as a discrete random variable Xi. The probability distribution Pr(
X
-
∞
+
∞
) over bi-infinite sequences
X
-
∞
+
∞
of these random variables is what we refer to as a stochastic process. Such a process is a complete description of a system’s behavior … Among all possible stochastic processes, several models have proven successful in order to model a wide range of practical situations. One such class of processes are the so-called Hidden Markov Models (HMM from now on). There are several possible ways to define an HMM. The most common among mathematicians is as a stochastic process S for which every symbol sn is generated conditionally from a Markovian process X, with probability Pr(sn|xn)" teaches that the input sequences represent temporal series measurements (input time series). Section 3, first paragraph: "We now turn to quantum systems and the process languages they generate. As with stochastic processes, we assume that the evolution of a quantum system is recorded as a sequence of measurement outcomes. Just like before, the distribution over sequences of these random variables is a process" teaches that the sequences for the quantum system are sequences of measurement outcomes (time series)), and
wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols (Section 2, first-third paragraphs: "Consider the temporal evolution of some natural system. The evolution is monitored by a series of measurements—numbers sequentially registered. Considering the numbers to be discrete, each such measurement can be taken as a discrete random variable Xi. The probability distribution Pr(
X
-
∞
+
∞
) over bi-infinite sequences
X
-
∞
+
∞
of these random variables is what we refer to as a stochastic process. Such a process is a complete description of a system’s behavior … Among all possible stochastic processes, several models have proven successful in order to model a wide range of practical situations. One such class of processes are the so-called Hidden Markov Models (HMM from now on). There are several possible ways to define an HMM. The most common among mathematicians is as a stochastic process S for which every symbol sn is generated conditionally from a Markovian process X, with probability Pr(sn|xn)" teaches that the input sequences represent temporal series measurements (input time series). Section 3, first paragraph: "We now turn to quantum systems and the process languages they generate. As with stochastic processes, we assume that the evolution of a quantum system is recorded as a sequence of measurement outcomes. Just like before, the distribution over sequences of these random variables is a process" teaches that the sequences for the quantum system are sequences of measurement outcomes (time series). Section 4.2, second paragraph: "A Hidden Quantum Markov Model (HQMM) is a d-level quantum system ρ together with a set of quantum operations Ks such that ΣsKs is trace-preserving. At every time step a symbol is generated with probability Pr(s) = tr[Ksρ] and the state vector is updated to ρs = Ksρ/Pr(s)" teaches that the HQMM quantum operation (e.g. computing kernel element) comprise generating a symbol at every time step (e.g. input time series is split into sequence of symbols)).
Bocharov et al., Markov et al., and Monras et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the two input sequences respectively represent two input time series, and wherein computing the kernel element comprises splitting the two input time series into the two sequences of symbols as taught by Monras et al. to the disclosed invention of Bocharov et al. in view of Markov et al.
One of ordinary skill in the art would have been motivated to make this modification because "the advantages of considering HQMMs are many-fold. On the one hand, it is sufficiently general to naturally encompass all Hidden Markov Models, a highly desirable feature for any quantum model attempting to outperform their classical counterpart. On the other hand, they provide the most compressed description of a quantum process by only accounting for the dynamics of the relevant part of the system, i.e., that which contains the quantum information latent in the system" (Monras et al. Section 7, first paragraph).
Claims 8 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Bocharov et al. (US 2021/0089953 A1) in view of Markov et al. ("Implementation and Learning of Quantum Hidden Markov Models") and further in view of Clark et al. ("Hidden Quantum Markov Models and Open Quantum Systems with Instantaneous Feedback").
Regarding Claim 8,
Bocharov et al. in view of Markov et al. teaches the system of claim 5.
Bocharov et al. in view of Markov et al. does not appear to explicitly teach wherein the quantum channel represents non restricted dynamics of an open quantum system that allows for information dissipation into an environment and information flow from the environment.
However, Clark et al. teaches wherein the quantum channel represents non restricted dynamics of an open quantum system that allows for information dissipation into an environment and information flow from the environment (Section 4, first paragraph: "Comparing the above description of open quantum systems with the definition of the HQMM in Sect. 2, it becomes relatively straightforward to see that open quantum systems with instantaneous feedback are concrete examples of HQMMs" teaches that the HQMM (quantum channel) represents an open quantum system. Section 1, fifth paragraph: "Like HQMMs, open quantum systems evolve randomly in time. Taking this perspective, the open quantum system itself provides the internal states of a HQMM, while its surrounding bath plays the role of the ancilla, which is constantly reset into an environmentally preferred, or einselected, state. By this, we mean the state that the environment would naturally evolve into if left alone. The continuous interaction between the internal states and the bath moves the bath away from its einselected state, thereby usually producing a measurable response that manifests itself as a random classical symbol. The effective dynamics of such a machine, when averaged over all possible trajectories, can be described by a Markovian master equation. When describing an open quantum system in this way, its accompanying output sequence is ignored. Here we suggest not to do so and to use the output sequences of open quantum systems to simulate stochastic processes" teaches that the HQMM (quantum channel) represents dynamics of an open quantum system that allows continuous interactions with the environment/bath (e.g. information dissipation and flow to and from the environment)).
Bocharov et al., Markov et al., and Clark et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the quantum channel represents non restricted dynamics of an open quantum system that allows for information dissipation into an environment and information flow from the environment as taught by Clark et al. to the disclosed invention of Bocharov et al. in view of Markov et al.
One of ordinary skill in the art would have been motivated to make this modification to "emphasize that open quantum systems with instantaneous feedback are examples of HQMMs, thereby identifying a novel application of quantum feedback control" (Clark et al. Abstract).
Regarding Claim 18,
Bocharov et al. in view of Markov et al. teaches the computer-implemented method of claim 13.
Bocharov et al. in view of Markov et al. does not appear to explicitly teach wherein the quantum channel represents non restricted dynamics of an open quantum system that allows for information dissipation into an environment and information flow from the environment.
However, Clark et al. teaches wherein the quantum channel represents non restricted dynamics of an open quantum system that allows for information dissipation into an environment and information flow from the environment (Section 4, first paragraph: "Comparing the above description of open quantum systems with the definition of the HQMM in Sect. 2, it becomes relatively straightforward to see that open quantum systems with instantaneous feedback are concrete examples of HQMMs" teaches that the HQMM (quantum channel) represents an open quantum system. Section 1, fifth paragraph: "Like HQMMs, open quantum systems evolve randomly in time. Taking this perspective, the open quantum system itself provides the internal states of a HQMM, while its surrounding bath plays the role of the ancilla, which is constantly reset into an environmentally preferred, or einselected, state. By this, we mean the state that the environment would naturally evolve into if left alone. The continuous interaction between the internal states and the bath moves the bath away from its einselected state, thereby usually producing a measurable response that manifests itself as a random classical symbol. The effective dynamics of such a machine, when averaged over all possible trajectories, can be described by a Markovian master equation. When describing an open quantum system in this way, its accompanying output sequence is ignored. Here we suggest not to do so and to use the output sequences of open quantum systems to simulate stochastic processes" teaches that the HQMM (quantum channel) represents dynamics of an open quantum system that allows continuous interactions with the environment/bath (e.g. information dissipation and flow to and from the environment)).
Bocharov et al., Markov et al., and Clark et al. are analogous to the claimed invention because they are directed towards quantum processing for sequential data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the quantum channel represents non restricted dynamics of an open quantum system that allows for information dissipation into an environment and information flow from the environment as taught by Clark et al. to the disclosed invention of Bocharov et al. in view of Markov et al.
One of ordinary skill in the art would have been motivated to make this modification to "emphasize that open quantum systems with instantaneous feedback are examples of HQMMs, thereby identifying a novel application of quantum feedback control" (Clark et al. Abstract).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN J HALES whose telephone number is (571)272-0878. The examiner can normally be reached M-F 9:00am - 5:00pm.
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, Kamran Afshar can be reached at (571) 272-7796. 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.
/BRIAN J HALES/Examiner, Art Unit 2125
/KAMRAN AFSHAR/Supervisory Patent Examiner, Art Unit 2125
Prosecution Insights