DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
2. Applicant's submission filed on 02 December 2025 [hereinafter Response] has been entered, where:
Claims 1-20 have been amended.
Claims 1-20 are pending.
Claims 1-20 are rejected.
Drawings
3. The objection to the drawings are as failing to comply with 37 CFR 1.84(p)(5) because they include the following reference character(s) not mentioned in the description is WITHDRAWN in view of the Applicant’s amendment to the drawings.
Claim Objections
4. The objection to claims 5, 7, 12, 14, and 19 because of informalities is WITHDRAWN in view of the Applicant’s amendments to the claims.
Claim Rejections – 35 U.S.C. § 112
5. The rejection to claims 6-14 and 20 under 35 U.S.C. § 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention is WITHDRAWN in view of the Applicant’s amendments to the claims.
Claim Rejections - 35 U.S.C. § 101
6. 35 U.S.C. § 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
7. Claims 1-20 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to an abstract idea without significantly more.
Claim 1 recites a computer-implemented method, which is a process, and thus one of the statutory categories of patentable subject matter. (35 U.S.C. § 101).
However, under Step 2A Prong One, the claim recites the limitations of “[(a.1)]1 generating, by the system, using quantum hardware, quantum output samples from a Hadamard circuit and a quantum imaginary-time evolution cooling,” and “[(a.2)] determining, by the system a Kullback-Leibler divergence gradient from the samples.” ” The limitations of “[(a.1)] generating” and “[(a.2)] determining” can practically be performed in the human mind, including, for example, observations, evaluations, judgments, and opinions, and accordingly, is a mental process, (MPEP §2106.04(a)(2) sub III), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Also, the limitations of [(a.1)] generating . . . quantum output samples . . . from a quantum imaginary-time evolution cooling” and “[(a.2)] determining”” are a mathematical concept, (MPEP § 2106.04(a)(2) sub I), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Thus, claim 1 recites an abstract idea.
Under Step 2A Prong Two, the claim as a whole is not integrated into a practical application, because the additional elements recited in the claim beyond the identified judicial exception include “computer-implemented,” a “system operatively coupled to a processor,” “quantum hardware,” and “Hadamard circuit.” These additional elements are recited at a high-level of generality, and are merely generic computer components used to implement the abstract idea, (MPEP § 2106.05(f)), that do not serve to integrate the abstract idea into a practical application. The claim also recites a “quantum Boltzmann machine learning model,” which is also recited at a high level of generality, and accordingly, is a generic computer component used to implement the abstract idea, (MPEP § 2106.05(f)), that does not integrate the abstract idea into a practical application. . The claim further recites the limitation of “[(a)] training, by a system operatively coupled to a processor, a quantum Boltzmann machine learning model,” which is the use of a generic computer components (computer-implemented, system, processor, quantum hardware, Hadamard circuit) to implement the abstract idea, (MPEP § 2106.05(f)), that does not serve to integrate the abstract idea into a practical application. Also, the claim recites the limitation of “[(a.3)] updating, by the system, a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient,” which is a post-processing, insignificant extra-solution activity of outputting a result to converge a gradient result, (MPEP § 2106.05(g)), that does not serve to integrate the abstract idea into a practical application. Therefore, claim 1 is directed to the abstract idea.
Finally, under Step 2B, the additional elements, taken alone or in combination, do not represent significantly more than the abstract idea itself. The additional elements include “computer-implemented,” and a “system operatively coupled to a processor,” “quantum hardware,” and “Hadamard circuit.” These additional elements are recited at a high-level of generality, and are merely generic computer components used to implement the abstract idea, (MPEP § 2106.05(f)), that do not amount to significantly more than the abstract idea. The claim also recites a “quantum Boltzmann machine learning model,” which is also recited at a high level of generality, and accordingly, is a generic computer component used to implement the abstract idea, (MPEP § 2106.05(f)), that does not amount to significantly more than the abstract idea. The claim further recites the limitation of “[(a)] training, by a system operatively coupled to a processor, a quantum Boltzmann machine learning model,” which is the use of a generic computer components (computer-implemented, system, processor, quantum hardware, Hadamard circuit) to implement the abstract idea, (MPEP § 2106.05(f)), that does not amount to significantly more than the abstract idea. Also, the claim recites the limitation of “[(a.3)] updating, by the system, a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient,” which is a well-understood, routine, and conventional activity of storing parameter data of a variable, (MPEP § 2106.05(d) sub II.iv), that does not amount to significantly more than the abstract idea. Therefore, claim 1 is subject-matter ineligible.
Claim 8 recites a system, which is a machine, and thus one of the statutory categories of patentable subject matter. (35 U.S.C. § 101).
However, under Step 2A Prong One, the claim recites the limitation of “[(a.1)] an evaluation component . . . generating , using the quantum hardware system, quantum output samples from a Hadamard circuit and a quantum imaginary-time evolution cooling,” and “[(a.2)] an evaluation component . . . determines, by the system a Kullback-Leibler divergence gradient from the samples.” ” The limitations of “[(a.1)] generating” and “[(a.2)] determining” can practically be performed in the human mind, including, for example, observations, evaluations, judgments, and opinions, and accordingly, is a mental process, (MPEP §2106.04(a)(2) sub III), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Also, the limitations of [(a.1)] generating . . . quantum output samples . . . from a quantum imaginary-time evolution cooling” and “[(a.2)] determine”” are a mathematical concept, (MPEP § 2106.04(a)(2) sub I), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Thus, claim 8 recites an abstract idea.
Under Step 2A Prong Two, the claim as a whole is not integrated into a practical application, because the additional elements recited in the claim beyond the identified judicial exception include a “quantum hardware system,” “memory that stores computer executable components,” a “processor,” and “Hadamard circuit.” These additional elements are recited at a high-level of generality, and are merely generic computer components used to implement the abstract idea, (MPEP § 2106.05(f)), that do not serve to integrate the abstract idea into a practical application. The claim also recites a “quantum Boltzmann machine learning model,” which is also recited at a high level of generality, and accordingly, is a generic computer component used to implement the abstract idea, (MPEP § 2106.05(f)), that does not integrate the abstract idea into a practical application. . The claim further recites the limitation of “[(a)] an evaluation component that trains a quantum Boltzmann machine learning model,” which is the use of a generic computer components (quantum hardware system, memory, processor, Hadamard circuit) ) to implement the abstract idea, (MPEP § 2106.05(f)), that does not serve to integrate the abstract idea into a practical application. Also, the claim recites the limitation of “[(b)] an update component that further trains the quantum Boltzmann machine learning model by [(b.1)] updating a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient,” which is a post-processing, insignificant extra-solution activity of outputting a result to converge a gradient result, (MPEP § 2106.05(g)), that does not serve to integrate the abstract idea into a practical application. Therefore, claim 8 is directed to the abstract idea.
Finally, under Step 2B, the additional elements, taken alone or in combination, do not represent significantly more than the abstract idea itself. The additional elements include a “quantum hardware system,” “memory that stores computer executable components,” a “processor,” and “Hadamard circuit.” These additional elements are recited at a high-level of generality, and are merely generic computer components used to implement the abstract idea, (MPEP § 2106.05(f)), that do not amount to significantly more than the abstract idea. The claim also recites a “quantum Boltzmann machine learning model,” which is also recited at a high level of generality, and accordingly, is a generic computer component used to implement the abstract idea, (MPEP § 2106.05(f)), that does not amount to significantly more than the abstract idea. The claim further recites the limitation of “[(a)] an evaluation component that trains a quantum Boltzmann machine learning model,” which is the use of a generic computer components (quantum hardware system, memory, processor, Hadamard circuit) ) to implement the abstract idea, (MPEP § 2106.05(f)), that does not amount to significantly more than the abstract idea. Also, the claim recites the limitation of “[(b)] an update component that further trains the quantum Boltzmann machine learning model by [(b.1)] updating a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient,” which is a well-understood, routine, and conventional activity of storing parameter data of a variable, (MPEP § 2106.05(d) sub II.iv), that does not amount to significantly more than the abstract idea. Therefore, claim 8 is subject-matter ineligible.
Claim 15 recites a non-transitory computer readable storage medium, which is a product, and thus is one of the statutory categories of patentable subject matter. (35 U.S.C. § 101).
However, under Step 2A Prong One, the claim recites the limitations of “[(a.1)] generate, by the processor, using quantum hardware, quantum output samples from a Hadamard circuit and a quantum imaginary-time evolution cooling,” and “[(a.2)] determine, by the processor, a Kullback-Leibler divergence gradient from the samples.” The limitations of “[(a.1)] generate” and “[(a.2)] determine” can practically be performed in the human mind, including, for example, observations, evaluations, judgments, and opinions, and accordingly, is a mental process, (MPEP §2106.04(a)(2) sub III), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Also, the limitations of [(a.1)] generate . . . quantum output samples . . . from a quantum imaginary-time evolution cooling” and “[(a.2)] determine” are a mathematical concept, (MPEP § 2106.04(a)(2) sub I), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Thus, claim 15 recites an abstract idea.
Under Step 2A Prong Two, the claim as a whole is not integrated into a practical application, because the additional elements recited in the claim beyond the identified judicial exception include “non-transitory computer readable storage medium having program instructions . . . executable by a processor,” “quantum hardware,” and “Hadamard circuit.” These additional elements are recited at a high-level of generality, and are merely generic computer components used to implement the abstract idea, (MPEP § 2106.05(f)), that do not serve to integrate the abstract idea into a practical application. The claim also recites a “quantum Boltzmann machine learning model,” which is also recited at a high level of generality, and accordingly, is a generic computer component used to implement the abstract idea, (MPEP § 2106.05(f)), that does not integrate the abstract idea into a practical application. . The claim further recites the limitation of “[(a)] training, by a system operatively coupled to a processor, a quantum Boltzmann machine learning model,” which is the use of a generic computer components (non-transitory computer readable storage medium, processor, quantum hardware, Hadamard circuit) to implement the abstract idea, (MPEP § 2106.05(f)), that does not serve to integrate the abstract idea into a practical application. Also, the claim recites the limitation of “[(a.3)] update, by the processor, a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient,” which is a post-processing, insignificant extra-solution activity of outputting a result to converge a gradient result, (MPEP § 2106.05(g)), that does not serve to integrate the abstract idea into a practical application. Therefore, claim 15 is directed to the abstract idea.
Finally, under Step 2B, the additional elements, taken alone or in combination, do not represent significantly more than the abstract idea itself. The additional elements include “non-transitory computer readable storage medium having program instructions . . . executable by a processor,” “quantum hardware,” and “Hadamard circuit.” These additional elements are recited at a high-level of generality, and are merely generic computer components used to implement the abstract idea, (MPEP § 2106.05(f)), that do not amount to significantly more than the abstract idea. The claim also recites a “quantum Boltzmann machine learning model,” which is also recited at a high level of generality, and accordingly, is a generic computer component used to implement the abstract idea, (MPEP § 2106.05(f)), that does not amount to significantly more than the abstract idea. The claim further recites the limitation of “[(a)] training, by a system operatively coupled to a processor, a quantum Boltzmann machine learning model,” which is the use of a generic computer components (non-transitory computer readable storage medium, processor, quantum hardware, Hadamard circuit) to implement the abstract idea, (MPEP § 2106.05(f)), that does not amount to significantly more than the abstract idea. Also, the claim recites the limitation of “[(a.3)] update, by the processor, a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient,” which is a well-understood, routine, and conventional activity of storing parameter data of a variable, (MPEP § 2106.05(d) sub II.iv), that does not amount to significantly more than the abstract idea. Therefore, claim 15 is subject-matter ineligible.
Claim 2 depends from claim 1. Claim 9 depends from claim 8. Claim 16 depends from claim 15. The claims recite more details or specifics to the additional element of the “Hadamard circuit,” (claims 2, 9, 16: “wherein the Hadamard circuit comprises a first unitary gate and a second unitary gate”), and accordingly, is merely more specific to the additional element. The abstract idea of these claims are not integrated into a practical application, (see MPEP § 2106.04(d)), nor do they amount to significantly more than the abstract idea, (MPEP § 2106.05 sub II), because the claims recite no more than the abstract idea. Therefore, claims 2, 9, and 16 are subject-matter ineligible.
Claim 3 depends directly or indirectly from claim 1. Claim 10 depends directly or indirectly from claim 8. Claim 17 depends directly or indirectly from claim 15. The claims recite the further limitation of ”[(b)] inputting . . . the Hamiltonian parameter configuration into the quantum imaginary-time evolution cooling and the Hadamard circuit.” Under Step 2A Prong Two, the limitation is an pre-processing, insignificant extra-solution activity of data input, (MPEP § 2106.05(g)), that does not serve to integrate the abstract idea into a practical application. Under Step 2B, the limitation is a well-understood, routine, and conventional activity of receiving or transmitting data over a network, (MPEP § 2106.05(d) sub II.i), that does not amount to significantly more than the abstract idea. Therefore, claims 3, 10, and 17 are subject-matter ineligible.
Claim 4 depends directly or indirectly from claim 1. Claim 11 depends directly or indirectly from claim 8. Claim 18 depends directly or indirectly from claim 15. The claims recite more details or specifics of the additional element of “(b)] updating,” (claims 3, 10, and 17: “wherein the Hamiltonian parameter configuration comprises angles of single qubit rotations”), and accordingly, are merely more specific to the additional element. the limitation of “[(c)] updating . . . the Hamiltonian parameter configuration to minimize the Kullback-Leibler divergence gradient.” Under Step 2A Prong Two, the limitation is an insignificant extra-solution activity of data input directed to increase accuracy of the samples, (MPEP § 2106.05(g)), that does not serve to integrate the abstract idea into a practical application. Under Step 2B, the limitation is a well-understood, routine, and conventional activity of storing or retrieving information in memory, (MPEP § 2106.05(d) sub II.iii), that does not amount to significantly more than the abstract idea. Therefore, claims 4, 11, and 18 are subject-matter ineligible.
Claim 5 depends directly or indirectly from claim 1. Claim 12 depends directly or indirectly from claim 8. Claim 19 depends directly or indirectly from claim 15. The claim recites more details or specifics of the abstract idea of the “[(a)] evaluating,” in that “[(a.2)] wherein the Kullback-Leibler divergence gradient is evaluated as:
PNG
media_image1.png
71
604
media_image1.png
Greyscale
,”
and accordingly, is merely more specific to the abstract idea. The abstract idea of these claims are not integrated into a practical application, (see MPEP § 2106.04(d)), nor do they amount to significantly more than the abstract idea, (MPEP § 2106.05 sub II), because the claims recite no more than the abstract idea. Therefore, claims 5, 12, and 19 are subject-matter ineligible.
Claim 6 depends directly or indirectly form claim 1. Claim 13 depends directly or indirectly from claim 8. Claim 20 depends directly or indirectly from claim 15. The claims recite the limitation of “[(b)] evaluating . . . a Kullback-Leibler divergence hessian of sample states.” The limitation of “[(b)] evaluating” can practically be performed in the human mind, including, for example, observations, evaluations, judgments, and opinions, and accordingly, is a mental process, (MPEP §2106.04(a)(2) sub III), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Also, the limitations of “[(a)] evaluating . . . a Kullback-Leibler divergence hessian” is a mathematical concept, (MPEP § 2106.04(a)(2) sub I), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). The abstract idea of these claims are not integrated into a practical application, (see MPEP § 2106.05(d)), nor do they amount to significantly more than the abstract idea, (MPEP § 2106.05(d)), because the claims recite no more than the abstract idea. Therefore, claims 6, 13, and 20 are subject-matter ineligible.
Claim 7 depends directly or indirectly from claim 1. Claim 14 depends directly or indirectly from claim 8. The claim recites more details or specifics of the abstract idea of the “[(b)] evaluating,” “[(b.1)] wherein the Kullback-Leibler divergence hessian is evaluated as:
PNG
media_image2.png
119
656
media_image2.png
Greyscale
. . . ,” and accordingly, is merely more specific to the abstract idea. The abstract idea of these claims are not integrated into a practical application, (see MPEP § 2106.04(d)), nor do they amount to significantly more than the abstract idea, (MPEP § 2106.05 sub II), because the claims recite no more than the abstract idea. Therefore claims 7 and 14 are subject-matter ineligible.
Claim Rejections – 35 U.S.C. § 103
8. 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.
9. 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.
10. 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.
11. Claims 1-4, 8-11, and 15-18 are rejected under 35 U.S.C. § 103 as being unpatentable over US Published Application 20210166133 to Ronagh et al. [hereinafter Ronagh] in view of Bauer et al., "Quantum algorithms for quantum chemistry and quantum materials science," arXiv (Jul 2020) [hereinafter Bauer] and Goto et al., “Boltzmann sampling from the Ising model using quantum heating of coupled nonlinear oscillators,” Nature (2018) [hereinafter Goto].
Regarding claims 1, 8, and 15, Ronagh teaches [a] computer-implemented method (Ronagh, Fig. 1, teaches a “system 100 comprises a digital computer 110 comprising a processing device 112 and a memory 114 comprising a computer program [(that is, a computer-implemented method)] of claim 1, [a] system (Ronagh, Fig. 1, teaches a “system 100” [(that is, a system)]”) of claim 8, and [a] computer program product (Ronagh ¶ 0116 teaches “the device comprises a storage device including, by way of non-limiting examples, CD-ROMs, DVDs, flash memory devices, magnetic disk drives, magnetic tapes drives, optical disk drives [(that is, a computer program product)]”) of claim 15, comprising:
[(a)] training, by a system operatively coupled to a processor, a quantum Boltzmann machine learning model (Ronagh, Fig. 1, teaches a quantum Boltzmann machine learning model trained by a training unit [Examiner annotations in dashed-line text boxes]:
PNG
media_image3.png
958
1076
media_image3.png
Greyscale
Ronagh ¶ 0127 teaches “quantum computer simulator 124 is represented by a stochastic framework and a reinforcement learning is used by the optional training unit 128 to improve parameters defining a generative machine learning model. In one or more embodiments, the generative machine learning model is a Restricted Boltzmann Machine [(that is, training, by a system operatively coupled to a processor, a quantum Boltzmann machine learning model)]”), wherein the training comprises:
* * *
Though Ronagh teaches mimicking a physics-inspired computer output for a given input, and updating at least one of said tunable parameters using said training unit to thereby improve a corresponding performance, Ronagh does not explicitly teach –
* * *
[(a) training a quantum Boltzmann machine learning model], wherein the training comprises:
[(a.1)] generating, by the system, using quantum hardware, quantum output samples from a Hadamard circuit and a quantum imaginary-time evolution cooling;
* * *
But Bauer teaches –
* * *
[(a) training a quantum Boltzmann machine learning model], wherein the training comprises:
[(a.1)] generating, by the system, using quantum hardware, quantum output samples (Bauer, left column of p. 25, “K. Finite-Temperature Algorithms,” first full paragraph, teaches “quantum Metropolis sampling samples from the Gibbs state in an analog of classical Metropolis sampling [(that is, quantum output samples)], using phase estimation on a random unitary applied to the physical qubits to ‘propose’ moves, and an iterative amplification procedure to implement the ‘rejection’”) from a Hadamard circuit (Bauer, Fig. 17, teaches a quantum circuit having a Hadamard gate [Examiner annotations in dashed-line text boxes]:
PNG
media_image4.png
252
488
media_image4.png
Greyscale
Bauer, Fig. 17 caption, teaches a “[q]uantum circuit for iterative [quantum phase estimation (QPE)]. The first two single-qubit gates bring the ancilla to the state
PNG
media_image5.png
40
173
media_image5.png
Greyscale
The controlled-U(t) gate, with
U
t
=
e
i
t
H
^
, brings the register into the state Eq. (16). The last Hadamard gate is needed to measure in the X basis) and a quantum imaginary-time evolution cooling (Bauer, left column of p. 16, “2. Quantum Imaginary-Time Evolution,” first paragraph, teaches “[i]n classical simulations, one popular approach to prepare (nearly exact) ground-states is imaginary-time evolution, which expresses the ground-state as the long-time limit of the imaginary-time Schrodinger equation,
PNG
media_image6.png
78
356
media_image6.png
Greyscale
Imaginary time-evolution underlies the family of projector quantum Monte Carlo methods in classical algorithms”; Bauer, Fig. 19, teaches energy as a function of inverse temperature β [Examiner annotations in dashed-line text boxes]:
PNG
media_image7.png
332
437
media_image7.png
Greyscale
Bauer, left column of p. 25, K. Finite-Temperature Algorithm,” first full paragraph, teaches a “quantum minimally entangled typical thermal state (METTS) algorithm [154] samples from the Gibbs state using imaginary time evolution applied to pure states, implemented via the quantum imaginary time evolution algorithm [(that is, a component of the “METTS algorithm,” having an increasing inverse temperature β, is a quantum imaginary-time evolution cooling)]”);
* * *
Ronagh and Bauer are from the same or similar field of endeavor. Ronagh teaches mimicking a physics-inspired computer output for a given input, and updating at least one of said tunable parameters using said training unit to thereby improve a corresponding performance. Bauer teaches performing a measurement on the final state and thus read out the result of the computation that is achieved through projective measurements of individual qubits in the computational basis in relation to a Hadamard circuit and quantum imaginary-time evolution cooling.
Thus, it would have been obvious to a person having ordinary skill in the art as of the effective filing date of the Applicant’s invention to modify Ronagh pertaining to training a Boltzmann engine with the quantum measurement sampling of Bauer.
The motivation to do so is because “[o]ne of the earliest and most compelling applications for quantum computers is Feynmans idea of simulating quantum systems with many degrees of freedom. Such systems are found across chemistry, physics, and materials science.” (Bauer, Abstract).
Though Ronagh and Bauer teach training a Boltzmann machine based on quantum data samples of a Hadamard circuit and quantum imaginary-time evolution cooling, the combination of Ronagh and Bauer, however, does not explicitly teach –
* * *
[(a) training a quantum Boltzmann machine learning model], wherein the training comprises:
* * *
[(a.2)] determining, by the system a Kullback-Leibler divergence gradient from the samples; and
[(a.3)] updating, by the system, a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient.
* * *
But Goto teaches –
* * *
[(a) training a quantum Boltzmann machine learning model], wherein the training comprises:
* * *
[(a.2)] determining, by the system , a Kullback-Leibler divergence gradient from the samples (Goto at p. 8, “Discussion,” third paragraph, teaches that “[s]ince the simulation of Boltzmann sampling from the Ising model is computationally hard in general for current digital computers, such a special-purpose machine for the Boltzmann sampling may be useful [(that is, a “digital computer” or a “special-purpose machine” is a system)]”; Goto at p. 8, “Methods,” first paragraph, teaches “k, the Boltzmann distribution is fitted to the simulation results of spin configuration probabilities by minimizing the Kullback-Leibler (KL) divergence DKL between them, where the KL divergence between two probability distributions {Pn} and {Qn} is defined as follows:
PNG
media_image8.png
68
530
media_image8.png
Greyscale
Note that the KL divergence is asymmetric with respect to the two distributions. In this paper, we choose the Boltzmann distribution as {Pn} and the simulation results as {Qn}”); and
[(a.3)] updating, by the system, a Hamiltonian parameter configuration of the quantum hardware to minimize the Kullback-Leibler divergence gradient (Goto at p. 2, “Quantum bifurcation machine for the Ising problem with local fields,” second paragraph, teaches “the extended [quantum bifurcation machine (QbM)] is defined by the following Hamiltonian in a frame rotating at half the pump frequency, ωp/2, of the parametric drive and in the rotating-wave approximation:
PNG
media_image9.png
172
776
media_image9.png
Greyscale
Goto at p. 3, “Quantum bifurcation machine for the Ising problem with local fields,” first full paragraph, teaches “To verify the validity of the above discussion, we numerically investigate an instance of two KPOs (N = 2), where the Schrödinger equation with the Hamiltonian in Eq. (2) is solved numerically. The time-dependent pump amplitude p(t) is increased linearly, as shown in Fig. 1a. The parameters of the instance [(that is, the “parameters” are a Hamiltonian parameter configuration of the quantum hardware)] are J1,2 = J2,1 = 1, h1 = −0.2, and h2 = 0, which are set such that two local minima exist in the energy landscape, as shown in Fig. 1b”; Goto, Fig. 2(d) teaches a minimized KL divergence gradient DKL [Examiner annotations in dashed-line text boxes]:
PNG
media_image10.png
405
336
media_image10.png
Greyscale
Goto, Fig. 2 caption teaches “Kullback-Leibler (KL) divergence DKL minimized for the fitting in [(c) (c) Inverse effective temperature β for the three values of pf determined by fitting to the instantaneous probability distribution]. (e and f) Time evolutions of β and DKL for various values of κ (pf = 4K)”).
[Examiner notes that the ordinary meaning of the term “Kullback-Leibler divergence gradient” a statistical measure that quantifies the difference between two probability distributions used to measure how one probability distribution diverges from a second, expected or target, probability distribution, where a “gradient” pertains to the rate and direction of change. Accordingly, the broadest reasonable interpretation of “Kullback-Leibler divergence gradient” covers the teachings of Goto, which is not inconsistent with the Applicant’s disclosure. (MPEP § 2111)]).
Ronagh, Bauer, and Goto are from the same or similar field of endeavor. Ronagh teaches mimicking a physics-inspired computer output for a given input, and updating at least one of said tunable parameters using said training unit to thereby improve a corresponding performance. Bauer teaches performing a measurement on the final state and thus read out the result of the computation that is achieved through projective measurements of individual qubits in the computational basis in relation to a Hadamard circuit and quantum imaginary-time evolution cooling. Goto teaches fitting the Boltzmann distribution to the simulation results, where the gradient of the Kullback-Leibler (KL) divergence DKL between the two distributions is minimized, where Boltzmann sampling is used for Boltzmann machine learning in the field of artificial intelligence.
Thus, it would have been obvious to a person having ordinary skill in the art as of the effective filing date of the Applicant’s invention to modify the combination of Ronagh and Bauer pertaining to training a Boltzmann engine with the quantum measurement sampling with the DKL of Goto.
The motivation to do so is because “[our] numerical study suggests that the output probability distribution of the dissipative QbM is Boltzmann-like, where the energy in the Boltzmann distribution corresponds to the cost function of the optimization problem.” (Goto, Abstract).
Regarding claims 2, 9, and 16, the combination of Ronagh, Bauer, and Goto teaches all of the limitations of claims 1, 8, and 15, respectively, as described above in detail.
Bauer teaches -
[(a.1.1)] wherein the Hadamard circuit comprises a first unitary gate and a second unitary gate (Bauer, Fig. 17, teaches samples from a Hadamard circuit [Examiner annotations in dashed-line text boxes]:
PNG
media_image11.png
242
442
media_image11.png
Greyscale
.
Bauer, Fig. 17 caption teaches “first two single [(that is, unitary)] qubit gates [(that is, a first unitary gate and a second unitary gate)] bring the ancilla to the state
PNG
media_image12.png
27
130
media_image12.png
Greyscale
. . . . The last Hadamard gate [(that is, a second unitary gate)] is needed to measure in the X basis”)
Regarding claims 3, 10, and 17, the combination of Goto and Bauer teach all of the limitations of claims 1, 8, and 15, respectively, as described above in detail.
Bauer teaches -
[(b)] inputting, by the system, the Hamiltonian parameter configuration into the quantum imaginary-time evolution cooling and the Hadamard circuit (Bauer, Fig. 14, teaches a classical optimization routine [Examiner annotations in dashed-line text boxes]:
PNG
media_image13.png
488
587
media_image13.png
Greyscale
Bauer, Fig. 14 caption, teaches a “classical optimization routine adds expectation values of the Hamiltonian Pauli terms to calculate the energy and estimates new values for the unitary parameters. The process is repeated until convergence [(that is, “optimization” is a quantum imaginary-time evolution)]”; Bauer, right column of p. 11, “B. Qubit Representation of Man-Body Systems,” first full paragraph, teaches “one can write down a simple form of the Hamiltonian a priori that contains the main interactions (a model Hamiltonian) with adjustable parameters; Bauer, right column of p. 17, “2. Hardware-Efficient Ansatz,” first paragraph, teaches “[g]iven a current [Hamiltonian] parameter configuration θ1 : : : θn [(that is, a Hamiltonian parameter configuration)], the commutator of the Hamiltonian with each operator in the pool is measured to obtain the gradient of the energy).
Regarding claims 4, 11, and 18, the combination of Goto and Bauer teach all of the limitations of claims 1, 8, and 15, respectively, as described above in detail.
Goto teaches –
[(a.3.1)] wherein the Hamiltonian parameter configuration comprises angles of single qubit rotations (Goto at p. 2, first paragraph, explains “quantum heating is the heating process induced by dissipation in quasienergy levels of driven dissipative quantum nonlinear systems, where the quasienergies are defined as eigenvalues of the system Hamiltonian in a rotating frame [(that is, “rotating frame” pertains to angles of single qubit rotations)]”; Goto at p. 2, “Quantum Bifurcation Machine for the Ising Problem with Local Fields,” first paragraph, teaches “the QbM to the Ising problem with local fields, which is to find the spin configuration that minimizes the following dimensionless Ising energy:
PNG
media_image14.png
80
337
media_image14.png
Greyscale
where si is the i-th Ising spin, which takes +1 (up) or −1 (down), N is the total number of Ising spins, s = (s1 s2 . . . sN) is the vector representation of a spin configuration, and {Ji,j} and {hi} are the dimensionless parameters [(that is, the Hamiltonian parameter configuration)] corresponding to the coupling coefficients and local fields, respectively [(that is, “coefficients” pertain to an imaginary number format, which is angles of single qubit rotations)]).
12. Claims 6, 13, and 20 are rejected under 35 U.S.C. § 103 as being unpatentable over US Published Application 20210166133 to Ronagh et al. [hereinafter Ronagh] in view of Bauer et al., "Quantum algorithms for quantum chemistry and quantum materials science," arXiv (Jul 2020) [hereinafter Bauer] and Goto et al., “Boltzmann sampling from the Ising model using quantum heating of coupled nonlinear oscillators,” Nature (2018) [hereinafter Goto], and Domenico et al., “Spectral Entropies as Information-Theoretic Tools for Complex Network Comparison,” arXiv (2016) [hereinafter Domenico].
Regarding claims 6, 13, and 20, the combination of Ronagh, Bauer, and Goto teach all of the limitations of claims 1, 8, and 15, respectively, as described above in detail.
Though Ronagh, Bauer, and Goto teach the features of training a Boltzmann machine and evaluating a KL divergence gradient of a quantum circuit, the combination of Ronagh, Bauer, and Goto, however, does not explicitly teach -
[(b)] evaluating, by the system, a Kullback-Leibler divergence hessian of sample states.
But Domenico teaches -
[(b)] evaluating, by the system, a Kullback-Leibler divergence hessian of the sample states (Domenico, left column of p. 6, “V. Generalized Quantum Divergences between Two Complex Networks,” second paragraph, teaches “the introduction of divergences (a.k.a. quantum relative entropy) in quantum information theory is foundational to the quest to understand differences between quantum states, quantum and classical information, the quantification of the thermodynamic cost of communication as well as optimal protocols to transfer information [(that is, “quantum states” is the sample states)]”; Domenico, right column of p. 6, “A. Maximum Likelihood Estimation and Model Selection,” first full paragraph, teaches “The covariance matrix corresponding to this estimation is the Fisher information matrix (see Appendix B), whose classical counterpart is equivalent to the Hessian of the Kullback-Leibler divergence and it is used to assess [(that is, evaluating)] the quality of the spectral likelihood estimate [(that is, evaluating, by the system, a Kullback-Leibler divergence hessian of the sample states)]”).
Ronagh, Bauer, and Goto are from the same or similar field of endeavor. Ronagh teaches mimicking a physics-inspired computer output for a given input, and updating at least one of said tunable parameters using said training unit to thereby improve a corresponding performance. Bauer teaches performing a measurement on the final state and thus read out the result of the computation that is achieved through projective measurements of individual qubits in the computational basis in relation to a Hadamard circuit and quantum imaginary-time evolution cooling. Goto teaches fitting the Boltzmann distribution to the simulation results, where the gradient of the Kullback-Leibler (KL) divergence DKL between the two distributions is minimized, where Boltzmann sampling is used for Boltzmann machine learning in the field of artificial intelligence. Domenico teaches the evaluation of information implicitly represented in its quantum states.
Thus, it would have been obvious to a person having ordinary skill in the art as of the effective filing date of the Applicant’s invention to modify the combination of Ronagh, Bauer, and Goto pertaining to training a Boltzmann engine with the quantum measurement sampling based on a DKL gradient with sampling with the KL divergence hessian evaluation of Domenico.
The motivation to do so is because, “by minimizing the Kullback-Leibler divergence between an observed network and a parametric network model, inference of model parameter(s) by means of maximum-likelihood estimation can be achieved and model selection can be performed appropriate information criteria. . . . Our results imply that spectral based statistical inference in complex networks results in demonstrably superior performance as well as a conceptual backbone, filling a gap towards a network information theory.” (Domenico, Abstract).
Response to Arguments
13. Examiner has fully considered Applicant’s arguments, and responds below accordingly.
35 U.S.C. § 101
14. Applicant submits, with respect to Step 2A Prong One, “the updated Subject Matter Eligibility guidelines clarifies groupings of subject matter included in abstract ideas. The analysis provided in Example 37, claim 2 . . . . Likewise, the subject claims recite elements involving executing quantum circuits utilized in a quantum machine learning model, sampling outputs from the execution of the quantum circuits, and updating parameters of the quantum circuits that cannot be practically applied or performed as a mental process in the human mind.” (Response at p. 11).
Examiner’s Response:
Examiner respectfully disagrees because the claims recite the limitation of at least “[(a.2)] determining, by the system a Kullback-Leibler divergence gradient from the samples.” The activity of “determining” can practically be performed in the human mind, including, for example, observations, evaluations, judgments, and opinions, and accordingly, is a mental process, (MPEP § 2106.04(a)(2) sub III), which is one of the groupings of abstract ideas. (MPEP § 2106.04(a)(2)). Also, in relation to a KL divergence gradient, the limitation is one of mathematical relationships, mathematical formulas or equations, and mathematical calculations, and accordingly, is also a mathematical concept, (MPEP § 2106.04(a)(2) sub I).
The second claim set of Example 37 recites, inter alia, a limitation based on a quantity of memory allocation, in which recites:
* * *
determining the amount of use of each icon using a processor that tracks how much memory has been allocated to each application associated with each icon over a predetermined period of time.”
* * *
The claim is not considered to not be an abstract idea under Step 2A Prong One because the “claimed step of determining the amount of use of each icon by tracking how much memory has been allocated . . . is not practically performed in the human mind.” (Subject Mater Eligibility Examples: Abstract Ideas, at p. 3).
In contrast, the example limitation does not provide such restraints as those set out by Example 37.
Accordingly, claims 1-20 are subject-matter ineligible as set out above in detail.
35 U.S.C. § 103
15. Applicant submits that “[w]hile Bauer et al. may disclose the use of a quantum imaginary-time evolution, Bauer et al. does not teach, disclose or suggest the use of a quantum imaginary-time evolution cooling process. In contrast, the instant specification recites the use of an imaginary-time evolution cooling process as cooling phase in qubit registers of a quantum computer as part of training process for a quantum Boltzmann machine. Accordingly, the cited references fail to teach, disclose, or suggest .. generating, by the system, using quantum hardware, quantum output samples from a Hadamard circuit and a quantum imaginary-time evolution cooling . . . (emphasis added) as recited in claim 1, and similarly in claims 8 and 15.” (Response at p. 13).
Examiner’s Response:
Examiner respectfully submits that the terminology referenced by Applicant are not tethered by the instant claims. For example, the claim simply recites “[(a.1)] generating, by the system, using quantum hardware, quantum output samples from a Hadamard circuit and a quantum imaginary-time evolution cooling.” (see claim 1, lines 4-6). In contrast, Applicant points to the specification for the proposition that the claim “recites the use of an imaginary-time evolution cooling process as cooling phase in qubit registers of a quantum computer as part of training process for a quantum Boltzmann machine.” The claims, however, are not so limited. Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 20 USPQ2d 1057 (Fed. Cir. 1993).
Also, Examiner notes that the “quantum imaginary-time evolution cooling” has a broadest reasonable interpretation of covering a “quantum imaginary-time evolution” because such an algorithm provides monitoring of the temperature characteristics of the process, which in quantum annealing, is a cooling process for determination of a ground state which is defined as a lowest energy state of the quantum system.
In this respect, the broadest reasonable interpretation of the term “quantum imaginary-time evolution cooling” covers the teachings of Bauer, (see at least Bauer, Fig. 19), as set out above in detail.
Conclusion
16. Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
17. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
(US Published Application 20140297247 to Troyer et al.) teaches quantum annealer simulator approximates unitary quantum dynamics of a quantum annealer on a non-quantum computing device such as a conventional computing device. The quantum annealer simulator may utilize algorithms that may efficiently approximate unitary time evolution of a quantum system, where the quantum system corresponds to a problem for which an optimized solution is sought.
(Silva et al., “Imaginary-Time Evolution Algorithms for Intermediate-Scale Quantum Signal Processors,” arXiv (2021)) teaches a new type of high-precision algorithms amenable to mid-term devices. First, we present two QITE primitives featuring excellent complexity scalings. In fact, one of them is optimal in ancillary overhead (requiring a single ancillary qubit throughout) whereas the other one is optimal in runtime for small inverse temperature for high precision ε-1. The latter is shown by noting that the runtime saturates a cooling-speed limit that is the imaginary-time counterpart of the no fast-forwarding theorem of real-time simulations, which we prove. Second, we present a master algorithm for deterministic QITE with an overall runtime asymptotically better than that of coherent approaches and the same hardware requirements as probabilistic ones, remarkably.
18. Any inquiry concerning this communication or earlier communications from the Examiner should be directed to KEVIN L. SMITH whose telephone number is (571) 272-5964. Normally, the Examiner is available on Monday-Thursday 0730-1730.
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, KAKALI CHAKI can be reached on 571-272-3719. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/K.L.S./
Examiner, Art Unit 2122
/KAKALI CHAKI/Supervisory Patent Examiner, Art Unit 2122
1 Examiner has added reference designations to the claim limitations for the limited purpose of SME evaluations.