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 .
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
The following title is suggested: “Training Machine Learning Models via Sequential Measurements in Quantum Cluster States”.
The disclosure is objected to because of the following informalities:
In ¶3…”superstitions of '0' and '1'” should be ‘superpositions of '0' and '1'’
In ¶20…”Using controller-Z gates…” should be ‘Using controlled-Z gates…’
In ¶37…”though” should be ‘through’
In ¶69…”perform” should be ‘performed’
Appropriate correction is required.
Claim Objections
Claim 9 is objected to because of the following informalities: “perform optimize” should be ‘perform optimization of’. Appropriate correction is required.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
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 1-20 are rejected under 35 USC 103 as being unpatentable over A measurement-based variational quantum eigensolver to Ferguson et al. (hereinafter Ferguson) in view of Quantum machine learning beyond kernel methods to Jerbi et al. (hereinafter Jerbi).
Per claim 1, Ferguson discloses A computer-implemented method (pg. 1, Abstract…measurement-based variational quantum eigensolver (MB-VQE)) comprising:
entangling a plurality of qubits to create a cluster state, wherein the plurality of qubits includes at least an input qubit, an output qubit, and at least one ancilla qubit (pg. 2... "the quantum part of a MB-VQE comprises an ansatz state |ψa⟩, a custom state, and a measurement prescription... The custom state is then created by expanding |ψa⟩ into a bigger graph state. This is done by decoration, i.e. by adding new vertices and connecting them to pre-existing sites in the ansatz state... the auxiliary qubits of the custom state are then measured, with the remaining ones constituting the output |ψout⟩", where the pre-existing sites of the ansatz state correspond to input qubits, the auxiliary qubits correspond to ancilla qubits, and the remaining ones correspond to output qubits), …;
performing sequential local measurements of the cluster state to generate a plurality of measurement outcomes (pg. 2... "single-qubit measurements are performed on auxiliary qubits... adaptive measurements must be performed in a specific order", which constitutes sequential local measurements generating measurement outcomes);
rotating at least one of the plurality of qubits according to the plurality of measurement outcomes and rotation parameters of the unitary operation, wherein the sequential local measurements and rotating at least one of the plurality of qubits transform an input state of the input qubit into an output state of the output qubit (pg. 2... "To make the computation deterministic, so called byproduct operators and adaptive measurements are required. The former applies X and Z operators to the output qubits depending on the measurement results, while the latter involves adapting the measurement bases R(θ) based on earlier measurement outcomes...", where the byproduct operators and adaptive measurement bases R(θ) with angles θ correspond to rotating qubits according to measurement outcomes and rotation parameters, transforming the input/ansatz state into the output state; Fig. 1…” measurements in rotated bases R(θ) transforms |ψa⟩ into the output state |ψout⟩”); and
training the machine learning model based on the output state of the output qubit (pg. 2... "The cost function to be fed into the classical side of the MB-VQE is then calculated from |ψout⟩ (e.g. its energy), with the angles θ of the rotated bases R(θ) being the variational parameters over which the optimization occurs", which describes optimizing/training a variational model based on the output state).
Ferguson does not expressly disclose, but with Jerbi teaches: the input qubit represents data among a training data set of a machine learning model represented by a unitary operation (Jerbi: pg. 1…teaches a quantum machine learning model where input data from a training data set is encoded into quantum states and processed by parametrized quantum circuits (unitary operations), "a supervised machine learning problem often reduces to the task of fitting a parametrized function also referred to as the machine learning model—to a set of previously labeled points, called a training set”; pg. 2...input data from the training data set is encoded into quantum states and processed by parametrized quantum circuits (unitary operations), “feature encoding unitary Uϕ: X -> F that maps input vectors x... to n-qubit quantum states ρ(x)…", where the n-qubit quantum states correspond to input qubits representing data among the training data set, and the parametrized quantum circuits correspond to the machine learning model represented by a unitary operation).
Ferguson and Jerbi are analogous art because they are from the same field of endeavor encompassing variational quantum algorithms and quantum computing.
Before the effective filing date of the claimed invention, it would have been obvious to a person having ordinary skill in the art to combine the measurement-based variational quantum circuit approach of Ferguson with the quantum machine learning models of Jerbi.
The suggestion/motivation for doing so would have been to provide the benefits of reduced gate sequence lengths and lower coherence time requirements for quantum machine learning tasks. Ferguson explicitly teaches that "any VQE can be translated into a MB-VQE" (Ferguson: pg. 3) and that MB-VQE is advantageous because it "avoids the requirement of performing long gate sequences, which is currently challenging due to error accumulation" (Ferguson: pg. 4). Since quantum machine learning models (as in Jerbi) utilize parametrized quantum circuits similar to VQEs, a person having ordinary skill in the art would have been motivated to implement the QML models of Jerbi using the MB-VQE framework of Ferguson to improve performance on noisy intermediate-scale quantum (NISQ) devices.
Per claim 2, Ferguson combined with Jerbi discloses claim 1, Ferguson further disclosing wherein training the machine learning model comprises optimizing the rotation parameters of the unitary operation (Ferguson: pg. 2... "with the angles θ of the rotated bases R(θ) being the variational parameters over which the optimization occurs", where optimizing the angles θ corresponds to optimizing the rotation parameters).
Per claim 3, Ferguson combined with Jerbi discloses claim 2, Ferguson further discloses wherein optimizing the rotation parameters comprises: measuring the output state of the output qubit to obtain an output (Ferguson: pg. 2... "The cost function to be fed into the classical side of the MB-VQE is then calculated from |ψout⟩", which intrinsically requires measuring the output state to obtain an output); using a cost function to determine a score associated with the output; and tuning the rotation parameters of the unitary operation based on the score (Ferguson: pg. 2... "The cost function to be fed into the classical side of the MB-VQE is then calculated from |ψout⟩ (e.g. its energy), with the angles θ of the rotated bases R(θ) being the variational parameters over which the optimization occurs", where calculating the cost function and optimizing the angles θ corresponds to determining a score and tuning the rotation parameters based on the score).
Per claim 4, Ferguson combined with Jerbi discloses claim 1, Ferguson further disclosing wherein entangling the plurality of qubits to create the cluster state comprises using a plurality of controlled-Z gates to entangle the plurality of qubits (Ferguson: pg. 2... "The main resource of MBQC are so-called graph states. Graphs as in Fig. 1 are stabilizer states (eigenstates with +1 eigenvalues) of the operators Sn = Xn ∏k Zk, where k refers to the vertices connected to site n", where the application of controlled-Z gates between connected vertices creates such graph/cluster states).
Per claim 5, Ferguson combined with Jerbi discloses claim 1, Ferguson further discloses wherein performing the sequential local measurement of the cluster state comprises: measuring the input qubit to obtain a first measurement outcome; and measuring a first ancilla qubit among the at least one ancilla qubit to obtain a second measurement outcome, wherein the first ancilla qubit is measured using a measurement angle that depends on the first measurement outcome and the first ancilla qubit succeeds the input qubit in the cluster state (Ferguson: pg. 2... "adaptive measurements are required... involves adapting the measurement bases R(θ) based on earlier measurement outcomes. Consequently, adaptive measurements must be performed in a specific order", which describes measuring a first qubit to obtain a first outcome, and measuring a succeeding qubit using a measurement angle/basis that depends on the earlier measurement outcome).
Per claim 6, Ferguson combined with Jerbi discloses claim 5. Ferguson further discloses wherein: in response to measuring the input qubit, the cluster state is reduced to a reduced cluster state that entangles a subset of the plurality of qubits including the at least one ancilla qubit and the output qubit (Ferguson: pg. 2... "Depending on the measurement outcomes, the system is probabilistically projected into different states", which describes reducing the cluster state upon measurement); and rotating at least one of the plurality of qubits comprises, in response to measuring the input qubit, rotating the subset of the plurality of qubits according to the first measurement outcome and a first rotation parameter of the unitary operation (Ferguson: pg. 2... "applies X and Z operators to the output qubits depending on the measurement results, while the latter involves adapting the measurement bases R(θ) based on earlier measurement outcomes", which describes rotating the remaining unmeasured qubits according to the measurement outcome and rotation parameters).
Per claim 7, Ferguson combined with Jerbi discloses claim 1, Ferguson further discloses wherein rotating at least one of the plurality of qubits comprises rotating a subset of the plurality of qubits that excludes the input qubit (Ferguson: pg. 2... "applies X and Z operators to the output qubits depending on the measurement results", where applying operators to the output qubits intrinsically excludes the already measured input/ancilla qubits).
Claims 8-14 are substantially similar in scope and spirit to claims 1-7. Therefore, the rejections of claims 1-7 are applied accordingly. Ferguson explicitly describe a "closed feedback loop between a classical computer and a quantum processor" (Ferguson: pg. 1), the classical computer logically requires software, code, and program instructions stored on a non-transitory computer-readable medium to execute the classical optimization algorithms (such as gradient descent) and to rapidly calculate the adaptive measurement angles required for the quantum hardware to maintain determinism.
Claims 15-20 are substantially similar in scope and spirit to claims 1-7. Therefore, the rejections of claims 1-7 are applied accordingly. Ferguson explicitly describes executing its measurement-based algorithm via a "closed feedback loop between a classical computer and a quantum processor" (Ferguson: pg. 1), being a system having both at least one processor and quantum hardware.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Patents and/or related publications are cited in the Notice of References Cited (Form PTO-892) attached to this action to further show the state of the art with respect to machine learning model training using measurements in quantum cluster states.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALAN CHEN whose telephone number is (571)272-4143. The examiner can normally be reached M-F 10-7.
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.
/ALAN CHEN/Primary Examiner, Art Unit 2125