Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of the Claims
Claims 1, 12, and 20 have been amended. Claims 8-9 and 18 have been cancelled. Claims 1-7, 10-17, and 19-20 are pending and have been considered by the Examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-7, 10-17, and 19-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Claims 1-7, 10-11 and 20 each recites a device comprising a processor (a product), and claims 12-17 and 19 each recites a method. A product and a method each falls within one of the four statutory categories of patent eligible subject matter.
CLAIM 1
Step 2A Prong 1: A corresponding plurality of training objective function values computed at the exact objective function from the training input states is a mathematical calculation. Instant specification paragraph [0035], lines 3-7 discloses performing the mathematical calculation.
Compute an estimated optimal state of the exact objective function using the trained
Computing the estimated optimal state includes: starting at an initial state, computing a preliminary estimated optimal state by performing a plurality of fast-step iterations of a Monte Carlo algorithm with respective fast-step acceptance probabilities that are determined based at least in part on the approximated objective function is a mathematical calculation. Instant specification paragraph [0027] discloses a mathematical formula for the fast-step acceptance probability.
Performing a correction iteration that has a correction-step acceptance probability determined based at least in part on respective values of the approximated objective function and the exact objective function computed at the preliminary estimated optimal state is a mathematical calculation. Instant specification paragraph [0030] discloses a mathematical formula for the correction-step acceptance probability. The claim recites abstract ideas.
Step 2A Prong 2: A computing device comprising: a processor configured to perform operations amounts to a generic computer component for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
Train a machine learning model using a training dataset that includes: a plurality of training input states of an exact objective function, and a corresponding plurality of training objective function values amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Using the trained machine learning model amounts to invoking the model merely as a tool to perform an existing process under MPEP 2106.05(f).
Output programmatical control instructions to one or more hardware devices based at least in part on the estimated optimal state amounts to mere data-gathering, an insignificant post-solution activity under MPEP 2106.05(g). The feature amounts to transmitting programmatic control instructions to a hardware device.
The additional elements as disclosed above, alone or in combination, do not integrate the abstract ideas into a practical application as they are mere generic computer functions as disclosed in combination with an insignificant post-solution activity that are implemented to perform the abstract ideas disclosed above. The claim is directed to an abstract idea.
Step 2B: A computing device comprising: a processor configured to perform operations amounts to a generic computer component for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
Train a machine learning model using a training dataset that includes: a plurality of training input states of an exact objective function, and a corresponding plurality of training objective function values amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Using the trained machine learning model amounts to invoking the model merely as a tool to perform an existing process under MPEP 2106.05(f).
Output programmatical control instructions to one or more hardware devices based at least in part on the estimated optimal state amounts to transmitting programmatic control instructions to a hardware device. This is analogous to transmitting data over a network, which the courts have recognized as a well-understood, routine, conventional activity under MPEP 2106.05(d)(II).
The additional elements as disclosed above, in combination with the abstract ideas, are not sufficient to amount to significantly more than the abstract ideas as they are mere generic computer functions as disclosed in combination with a well-understood, routine, conventional activity that are implemented to perform the abstract ideas disclosed above. The claim is not patent eligible.
CLAIM 2 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. The Monte Carlo algorithm is a Markov chain Monte Carlo (MCMC) algorithm selected from the group consisting of a Metropolis-Hastings algorithm, a simulated annealing algorithm, a simulated quantum annealing algorithm, a parallel tempering algorithm, and a population annealing algorithm is a mathematical calculation. Specification paragraphs [0024]-[0025] disclose formulas for a Metropolis-Hastings algorithm.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas.
CLAIM 3 incorporates the rejection of claim 2.
Step 2A Prong 1: The abstract ideas of claim 2 are incorporated. Each of the fast-step iterations of the MCMC algorithm has a fast-step acceptance probability given by
PNG
media_image1.png
46
384
media_image1.png
Greyscale
where x is a current state, x' is an updated state, β is an inverse temperature, and
PNG
media_image2.png
30
38
media_image2.png
Greyscale
is a change in a value of the approximated objective function between the current state and the updated state is a mathematical calculation.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas.
CLAIM 4 incorporates the rejection of claim 3.
Step 2A Prong 1: The abstract ideas of claim 3 are incorporated. The correction-step acceptance probability of the correction iteration is given by
PNG
media_image3.png
48
458
media_image3.png
Greyscale
where ΔE is a change in a value of the exact objective function between the initial state and the preliminary estimated optimal state is a mathematical calculation.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas.
CLAIM 5 incorporates the rejection of claim 2.
Step 2A Prong 1: The abstract ideas of claim 2 are incorporated. The respective fast-step acceptance probabilities of the plurality of fast-step iterations are determined based at least in part on a constraint function in addition to the approximated objective function is a mathematical calculation. Specification paragraph [0032] discloses a formula for computing a fast-step acceptance probabilities based on a constraint function.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas.
CLAIM 6 incorporates the rejection of claim 5.
Step 2A Prong 1: The abstract ideas of claim 5 are incorporated. Each of the fast-step iterations of the MCMC algorithm has a fast-step acceptance probability given by
PNG
media_image4.png
50
434
media_image4.png
Greyscale
where x is a current state, x' is an updated state, β is an inverse temperature,
PNG
media_image2.png
30
38
media_image2.png
Greyscale
is a change in a value of the approximated objective function between the current state and the updated state, γ is a constraint function weighting parameter, and ΔC is a change in a value of the constraint function between the current state and the updated state is a mathematical calculation.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas.
CLAIM 7 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. The limitation “repeat an estimation loop that includes the plurality of fast-step iterations and the correction iteration until the correction iteration is accepted” amounts to repeating mathematical calculations.
Step 2A Prong 2 and Step 2B: A processor amounts to a generic computer component for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
CLAIM 10 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. During each of the fast-step iterations of the Monte Carlo algorithm, the processor is configured to sample from a Gibbs distribution over an approximated state space of the approximated objective function is a mathematical calculation. Instant specification paragraph [0023] discloses a mathematical formula for the Gibbs distribution.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas.
CLAIM 11 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. The Monte Carlo algorithm is a non-Markovian Monte Carlo algorithm in which the processor is configured to compute the preliminary estimated optimal state based at least in part on a sequence of one or more prior states is a mathematical calculation. Instant specification paragraph [0039] discloses a mathematical formula for a non-Markovian Monte Carlo algorithm.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas.
Claims 12-15 each recites a method which implements the same features as product claims 1-4, respectively, and are therefore rejected for at least the same reasons.
Claim 16 recites a method which implements the same features as product claims 5 and 6 and are therefore rejected for at least the same reasons.
Claims 17 and 19 each recites a method which implements the same features as product claims 7 and 11, respectively, and are therefore rejected for at least the same reasons.
CLAIM 20
Step 2A Prong 1: A corresponding plurality of training objective function values computed at the exact objective function from the training input states is a mathematical calculation. Instant specification paragraph [0035], lines 3-7 discloses performing the mathematical calculation.
Compute an estimated optimal state of the exact objective function using the trained
Computing the estimated optimal state includes, in one or more iterations of an estimation loop that includes a plurality of fast-step iterations and a correction iteration and that is repeated until the correction iteration is accepted are mathematical calculations. Instant specification paragraphs [0027] and [0030] disclose mathematical formulas for a fast-step acceptance probability and a correction-step acceptance probability, respectively.
Starting at an initial state, computing a preliminary estimated optimal state by performing the plurality of fast-step iterations, wherein: each of the fast-step iterations is an iteration of a Markov chain Monte Carlo (MCMC) algorithm with a respective fast-step acceptance probability that is determined based at least in part on the approximated objective function is a mathematical calculation. Instant specification paragraph [0027] discloses a mathematical formula for the fast-step acceptance probability.
The MCMC algorithm is selected from the group consisting of a Metropolis-Hastings algorithm, a simulated annealing algorithm, a simulated quantum annealing algorithm, a parallel tempering algorithm, and a population annealing algorithm is a mathematical calculation. Specification paragraphs [0024]-[0025] disclose formulas for a Metropolis-Hastings algorithm.
Performing the correction iteration, wherein the correction iteration is an iteration of the MCMC algorithm that has a correction-step acceptance probability determined based at least in part on respective values of the approximated objective function and the exact objective function computed at the preliminary estimated optimal state is a mathematical calculation. Instant specification paragraph [0030] discloses a mathematical formula for the correction-step acceptance probability. The claim recites abstract ideas.
Step 2A Prong 2: A computing device comprising: a processor amounts to a generic computer component for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
Train a machine learning model using a training dataset that includes: a plurality of training input states of an exact objective function, and a corresponding plurality of training objective function values amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Using the trained machine learning model amounts to invoking the model merely as a tool to perform an existing process under MPEP 2106.05(f).
Output programmatical control instructions to one or more hardware devices based at least in part on the estimated optimal state amounts to mere data-gathering, an insignificant post-solution activity under MPEP 2106.05(g). The feature amounts to transmitting programmatic control instructions to a hardware device.
The additional elements as disclosed above, alone or in combination, do not integrate the abstract ideas into a practical application as they are mere generic computer functions as disclosed in combination with an insignificant post-solution activity that are implemented to perform the abstract ideas disclosed above. The claim is directed to an abstract idea.
Step 2B: A computing device comprising: a processor amounts to a generic computer component for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
Train a machine learning model using a training dataset that includes: a plurality of training input states of an exact objective function, and a corresponding plurality of training objective function values amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Using the trained machine learning model amounts to invoking the model merely as a tool to perform an existing process under MPEP 2106.05(f).
Output programmatical control instructions to one or more hardware devices based at least in part on the estimated optimal state amounts to transmitting programmatic control instructions to a hardware device. This is analogous to transmitting data over a network, which the courts have recognized as a well-understood, routine, conventional activity under MPEP 2106.05(d)(II).
The additional elements as disclosed above, in combination with the abstract ideas, are not sufficient to amount to significantly more than the abstract ideas as they are mere generic computer functions in combination with a well-understood, routine, conventional activity as disclosed that are implemented to perform the abstract ideas disclosed above. The claim is not patent eligible.
Response to Arguments
The following is the Examiner’s response to the Applicant’s arguments filed 05/26/2026.
Applicant’s First Arguments Under 35 U.S.C. 101: Page 11 argues that by reducing the size of the search space over which the processor searches for the estimated optimal state, the computation of the preliminary estimated optimal state may converge more quickly compared to performing a Monte Carlo search with the full objective function. Since reducing the number of variables reduces the size of the search space and allows the Monte Carlo search to converge to the preliminary estimated optimal state more quickly, the feature of [previous] claim 8 recites an improvement to the functioning of the computing system and integrates any alleged abstract ideas into the practical application of Monte Carlo search. The feature "thereby reducing a search space size of the computation of the estimated optimal state" clarifies this technical effect.
Examiner’s Response: Applicant's arguments have been fully considered but they are not persuasive. In claim 1, the limitation “compute an estimated optimal state of the exact objective function using the trained… model as an approximated objective function, wherein: the approximated objective function has a reduced number of variables relative to the exact objective function, thereby reducing a search space size of the computation of the estimated optimal state” is a mathematical calculation. Instant specification paragraph [0020] discloses generating an approximated objective function by excluding terms from an exact objective function. Paragraph [0027] discloses a mathematical formula for a fast-step acceptance probability as a function of the approximated objective function, and paragraph [0030] discloses a mathematical formula for a correction-step acceptance probability as a function of the exact objective function.
A mathematical calculation cannot provide a technical improvement. MPEP 2106.05(a) states, “It is important to note, the judicial exception alone cannot provide the improvement.” MPEP 2106.05(a), subsection II. states, “it is important to keep in mind that an improvement in the abstract idea itself (e.g. a recited fundamental economic concept) is not an improvement in technology.”
Applicant’s Second Arguments Under 35 U.S.C. 101: Pages 11-12 argue that the Office action does not cite any references to support the assertion that training the machine learning model as in claim 1, followed by using the trained machine learning model to approximate the objective function, recites an existing process. Applicant instead respectfully submits that claim 1 recites an unconventional training data generation process that allows the trained machine learning model to be used in an unconventional operation. The trained machine learning model therefore does not correspond to a generic computer. In addition, approximating the objective function using the trained machine learning model achieves an improvement in the functioning of the computing device itself by allowing the computing device to efficiently obtain the estimated optimal state even in examples in which evaluating the exact objective function is computationally expensive, as disclosed in Para. [0051] of the subject application. Applicant therefore respectfully submits that the features of claim 1 are integrated into a practical application.
Examiner’s Response: Applicant's arguments have been fully considered but they are not persuasive. Applicant appears to have mischaracterized the Office action, because the Office action did not state that the entire limitation of claim 1, lines 3-6 is an additional element, nor did the Office action state that the entire limitation “compute an estimated optimal state of the exact objective function using the trained machine learning model as an approximated objective function” is an additional element. In Step 2A Prong 1 of the Office Action, the limitations “a corresponding plurality of training objective function values computed at the exact objective function from the training input states” and “compute an estimated optimal state of the exact objective function using the trained… model as an approximated objective function” are mathematical calculations. In Step 2A Prong 2, the limitation “train a machine learning model using a training dataset that includes: a plurality of training input states of an exact objective function, and a corresponding plurality of training objective function values” is an additional element which amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f). The limitation “using the trained machine learning model” is an additional element which amounts to invoking the model merely as a tool to perform an existing process under MPEP 2106.05(f).
In response to the Applicant’s argument that the Office action does not cite any references, an Office action should cite Berkheimer evidence when an additional element is an insignificant extra-solution activity in Step 2A Prong 2 and the additional element is not analogous to any well-understood, routine, and conventional computer functions identified by the courts in MPEP 2106.05(d) subsection II. In Step 2A Prong 2, the limitations of training a machine learning model and using the trained machine learning model as recited in claim 1 amount to mere instructions to apply the abstract ideas on a generic computer and invoking the model merely as a tool to perform an existing process under MPEP 2106.05(f). They do not cite Berkheimer evidence since they are not insignificant extra-solution activity in Step 2A Prong 2.
Examiner respectfully disagrees with the argument that the addition elements recited above are unconventional. The training process, the training dataset, and using the trained machine learning model are recited at a high level of generality. The training dataset includes “training input states” as training inputs and “training objective function values” as target outputs. Training a machine learning model on pairs of inputs and target values is a conventional process in the field of machine learning. Using a trained machine learning model to predict outputs on unseen input data is also a conventional process in the field of machine learning.
In Step 2A Prong 1, the limitations “computing an estimated optimal state of the exact objective function using the trained… model as an approximated objective function, wherein: the approximated objective function has a reduced number of variables relative to the exact objective function, thereby reducing a search space size of the computation of the estimated optimal state” is a mathematical calculation as disclosed in instant specification paragraphs [0020], [0027], [0030]. MPEP 2106.05(a) and subsection II state that mathematical calculations cannot provide a technical improvement.
Conclusion
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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Asher H. Jablon whose telephone number is (571)270-7648. The examiner can normally be reached Monday - Friday, 9:00 am - 6:00 pm.
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, Abdullah Al Kawsar can be reached at (571)270-3169. 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.
/A.H.J./Examiner, Art Unit 2127
/ABDULLAH AL KAWSAR/Supervisory Patent Examiner, Art Unit 2127
Prosecution Insights