DETAILED ACTION
This Office Action is in response to the amendments filed on 04/03/2026.
Claims 1 and 15 are currently amended.
Claims 11-14 and 19 are currently canceled.
Claims 20 and 21 are newly added.
Claims 1-10, 15-18, 20, and 21 are currently pending in this application and have been examined.
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 .
Response to Arguments
In reference to Applicant’s arguments on page(s) 10 regarding rejections made under 35 U.S.C. 112:
In the Office Action, claims 11-14 were rejected under 35 U.S.C. § 112(b) as being indefinite. In this response, claims 11-14 have been canceled. Accordingly, the rejections of claims 11-14 under 35 U.S.C. § 112(b) have been overcome and should be withdrawn.
Moreover, based on the cancelation of claims 11-14, it is understood that none of the presently recited claim terms invoke the means plus function limitations of 35 U.S.C. § 112(f).
Examiner’s response:
Applicant’s arguments have been fully considered and are found to be persuasive.
Claims 11-14, previously rejected under 35 U.S.C. 112(b), have been canceled and therefore the rejections made under 35 U.S.C 112 are withdrawn.
In reference to Applicant’s arguments on page(s) 10-12 regarding rejections made under 35 U.S.C. 101:
In the Office Action, claims 1-18 were rejected under 35 U.S.C. § 101 as allegedly being directed to an abstract idea without significantly more. In this response, independent claims 1 and 15 have been amended to recite a practical application of the claimed invention even more clearly. Reconsideration is respectfully requested.
As described in the specification, conventional approaches to score matching are inefficient and computationally burdensome for high-dimensional data, and marginal and posterior probability distributions are often intractable. (Specification at paragraphs [0029], [0033], and [0041]). The claimed method addresses these technical challenges using the BiSM method to approximate the true posterior probability distribution and to enable optimization based on an obtained marginal probability distribution without requiring a direct computation of these otherwise intractable quantities. Therefore, the claimed approach improves the functioning of the computer performing the training because the training is performed faster, requires less processing power, is more energy efficient, and overcomes the challenges of being unable to determine certain probability distributions.
The amended claims improve the functioning of a computer by enabling more efficient training of neural networks based on energy-based model with high-dimensional data using finite computational resources. These limitations impose meaningful constraints on any alleged abstract idea and integrate any alleged abstract idea into a practical application. At least some of the training challenges imposed by the latent variable are overcome.
For at least the above reasons, it is respectfully submitted that the 35 U.S.C. § 101 rejections of claims 1-10 and 15-18 have been overcome and should be withdrawn.
Examiner’s response:
Applicant’s arguments have been fully considered but are found to be not persuasive.
Applicant argues that the instant application presents a technological improvement to the operation of a computer because the use of a different method to calculate the true posterior probability distribution. Examiner disagrees. While the use of the different calculation method may lead to improved training (i.e. performed faster, requires less processing power, is more energy efficient), a technological improvement cannot stem from an abstract idea, in this case calculating approximate statistical distributions.
In light of the amendments made on the claims, the rejections made under 35 U.S.C. 101 are maintained and updated below.
In reference to Applicant’s arguments on page(s) 12-15 regarding rejections made under 35 U.S.C. 103:
In the Office Action, claims 1, 5-7, 11, and 15 were rejected under 35 U.S.C. § 103 as being unpatentable over U.S. Pat. Pub. No. US2006/0034495 Al (hereinafter "Miller"), in view of "Bayesian model reduction" (hereinafter "Friston"), in view of "Sliced Score Matching: A Scalable Approach to Density and Score Estimation" (hereinafter "Song"), and further in view of "Cooperative Learning of Energy-Based Model and Latent Variable Model via MCMC Teaching" (hereinafter "Xie"). Reconsideration is respectfully requested.
The Examiner alleged that Miller in view of Friston and Xie discloses the method of claim 1. (Office Action at page 26). Claim 1 has been amended to more clearly recite the claimed invention. The proposed combination does not arrive at the invention of claim 1.
In the rejection of claim 1, it was acknowledged that Miller fails to disclose the claimed marginal probability distribution and Friston and Xie were cited to account for this deficiency. (Office Action at page 28).
Friston was alleged to disclose a marginal probability distribution at page 5. (Office Action at page 28). The cited passage discloses a process of factorizing marginals from the approximate posterior (Q(theta)). Factorizing marginals from the approximate posterior does not correspond to the approach of claim 1 for approximating the marginal probability distribution. No other approach for the marginals is provided in Friston. Friston does not provide much disclosure regarding the marginals.
Xie was alleged to disclose the unnormalized joint probability distribution. (Office action at page 31).
It is respectfully submitted that the method of claim 1 is not disclosed by the combination of Xie and Friston. That is, a skilled person would not have combined the joint density model of Xie with the factorizing marginals approach of Friston to arrive at the claimed approach of determining the obtained marginal probability distribution based on (i) the variational posterior probability distribution, and (ii) an unnormalized joint probability distribution of the visible variable and the latent variable. That is, the references do not "connect" the approximation of the marginal probability distribution with an unnormalized joint probability distribution. At best, the skilled person would have arrived at a different approach for determining the evidence lower bound (ELPO) of Friston, which does not correspond to the approximate marginal probability distribution, as required by claim 1. As noted, Miller was not alleged to disclose and does not disclose this aspect of the claimed invention.
Therefore, even if Miller is modified based upon the teachings of Friston and Xie, the modification does not arrive at the method of claim 1. Accordingly, the Office Action has not made a prima facie case of obviousness with respect to claim 1, and the rejection thereof should be withdrawn.
Examiner’s response:
Applicant’s arguments have been fully considered and are found to be persuasive.
Applicant argues that the applied prior art reference of Miller in view of Friston and Xie does not disclose the method of Claim 1. Examiner agrees. The method(s) performed in Miller in view of Friston and Xie was mischaracterized as interpreting factorizing marginals from the approximate posterior to teach the step of approximating the true marginal probability distribution, this is incorrect as the mathematical steps taken are similar in naming conventions but not in application or results.
Applicant argues that the use of Xie is not appropriate. Examiner agrees. Xie was relied upon to teach the use of an unnormalized joint probability distribution, but one skilled in the art would not reasonably connect the two references in any meaningful way to bridge the gap between them.
In light of the arguments presented, the rejections made under 35 U.S.C. 103 are withdrawn.
Claim Rejections - 35 USC § 101
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1-10, 15-18, 20, and 21 rejected under 35 U.S.C. 101 because they are directed to an abstract idea without significantly more.
Step 1 analysis:
Independent Claim 1 recites, in part, a method, therefore falling into the statutory category of process. Independent Claims 11 and 15 recite, in part, an apparatus, therefore falling into the statutory category of machine.
Regarding Claim 1:
Step 2A: Prong 1 analysis:
Claim 1 recites in part:
“obtaining a variational posterior probability distribution of the latent variable given the visible variable by optimizing a set of parameters of the variational posterior probability distribution resulting from the minibatch of the training data”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical relationship/concept.
“wherein the variational posterior probability distribution is provided to approximate a true posterior probability distribution of the latent variable given the visible variable, and wherein the true posterior probability distribution is relevant to network parameters of the set of network parameters stored in the memory”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical relationship.
“optimizing the network parameters stored in the memory using the processor based on a score matching objective of an obtained marginal probability distribution on the minibatch of training data”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper. For example, this limitation encompasses optimizing parameters based on a calculated score, which, because there is no mention of a training step, could be as simple as choosing parameters that would have the biggest effect on the output.
“wherein the obtained marginal probability distribution is obtained based on the variational posterior probability distribution and an unnormalized joint probability distribution of the visible variable and the latent variable”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical relationship.
“wherein the obtained marginal probability distribution approximates a true marginal probability distribution”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical relationship.
“repeating the steps of providing the minibatch of the training data, obtaining the variational posterior probability distribution, and optimizing the network parameters on different minibatches of the training data, until a convergence condition is satisfied indicating that the neural network has been trained to perform the task”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper, and also covers a mathematical relationship, as determined in the above limitations. For example, this limitation encompasses repeating steps of calculating mathematical values until a condition is satisfied.
Accordingly, at Step 2A: Prong 1, the claim is directed to an abstract idea.
Step 2A: Prong 2 analysis:
The judicial exception is not integrated into practical application. In particular, the claim recites the additional elements of:
“using a processor”. This additional element is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a generic computer component (processor) (See MPEP 2106.05(f)).
“training the neural network, using the processor, to perform the task by”. This additional element is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a generic computer component (neural network) (See MPEP 2106.05(f)).
“wherein obtaining the variational posterior probability distribution and optimizing the network parameters based on the score matching objective overcomes intractability of the true marginal probability distribution and intractability of the true posterior probability distribution”. This additional element is recited at a high level of generality such that the claim recites only the idea of a solution or outcome (overcoming intractability) i.e., the claim fails to recite details of how a solution to a problem is accomplished.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
As discussed above, the additional element(s) of “using a processor” and “training the neural network, using the processor, to perform the task by” is/are recited at a high-level of generality such that it/they amount(s) to no more than mere instructions to apply the exception using generic computer components (processor and neural network) (See MPEP 2106.05(f)).
As discussed above, the additional element(s) of “wherein obtaining the variational posterior probability distribution and optimizing the network parameters based on the score matching objective overcomes intractability of the true marginal probability distribution and intractability of the true posterior probability distribution” is/are recited at a high-level of generality such that the claim recites only the idea of a solution or outcome (overcoming intractability) i.e., the claim fails to recite details of how a solution to a problem is accomplished (See MPEP 2106.05(f)).
Accordingly, at Step 2B, the additional elements individually or in combination do not amount to significantly more than the judicial exception.
Regarding Claim 2:
Step 2A: Prong 1 analysis:
Claim 2 recites in part:
“repeating following steps for a number of K times, wherein K is an integer equal to or greater than zero: calculating a stochastic gradient of the divergence objective under given network parameters”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical calculation.
“updating the set of parameters based on the calculated stochastic gradient by starting from an initialized or previously updated set of parameters”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical calculation.
Accordingly, at Step 2A: Prong 1, the claim is directed to an abstract idea.
Step 2A: Prong 2 analysis:
The claim does not recite any additional elements that integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
Regarding Claim 3:
Step 2A: Prong 1 analysis:
Claim 3 recites in part:
“calculating the set of parameters as a function of the network parameters recursively for a number of N times by starting from an initialized or previously updated set of parameters, wherein N is an integer equal to or greater than zero”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical calculation.
“obtaining an approximated stochastic gradient of the score matching objective based on the calculated set of parameters”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical calculation.
“updating the network parameters based on the approximated stochastic gradient”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical calculation.
Accordingly, at Step 2A: Prong 1, the claim is directed to an abstract idea.
Step 2A: Prong 2 analysis:
The claim does not recite any additional elements that integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
Regarding Claim 4:
Step 2A: Prong 1 analysis:
Claim 4 recites in part:
“wherein the variational posterior probability distribution is a Bernoulli distribution parameterized by a fully connected layer with sigmoid activation or a Gaussian distribution parameterized by a convolutional neural network”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical concept.
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2A: Prong 2 analysis:
The claim does not recite any additional elements that integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
Regarding Claim 5:
Step 2A: Prong 1 analysis:
Claim 5 recites in part:
“wherein optimizing the set of parameters of the variational posterior probability distribution is performed based on an objective of minimizing Kullback-Leibler divergence or Fisher divergence between the variational posterior probability distribution and the true posterior probability distribution”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper. For example, this limitation encompasses optimizing parameters based on a minimized score, which, because there is no mention of a training step, could be as simple as choosing parameters that would have the biggest effect on the output.
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2A: Prong 2 analysis:
The claim does not recite any additional elements that integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
Regarding Claim 6:
Step 2A: Prong 1 analysis:
Claim 6 recites in part:
“wherein the score matching objective is based at least in part on one of sliced score matching, denoising score matching, or multiscale denoising score matching”. As drafted and under its broadest reasonable interpretation, this limitation covers a mathematical concept.
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2A: Prong 2 analysis:
The claim does not recite any additional elements that integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
Regarding Claim 7:
Step 2A: Prong 2 analysis:
The judicial exception is not integrated into practical application. In particular, the claim recites the additional elements of:
“wherein the training data comprises at least one of image data, video data, and audio data”. This limitation merely indicates a field of use or technological environment in which the judicial exception is performed (data types) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h).
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
The additional element(s) of “wherein the training data comprises at least one of image data, video data, and audio data” is/are directed to particular field(s) of use (data types) (MPEP 2106.05(h)) and therefore do not provide significantly more than the abstract idea, and thus the claim is subject-matter ineligible.
Accordingly, at Step 2B, the additional elements individually or in combination do not amount to significantly more than the judicial exception.
Regarding Claim 8:
Step 2A: Prong 1 analysis:
Claim 8 recites in part:
“identifying the component to be detected as an abnormal component, if the density value is below a threshold”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper. For example, this limitation encompasses identifying an anomaly based on a threshold value.
Accordingly, at Step 2A: Prong 1, the claim is directed to an abstract idea.
Step 2A: Prong 2 analysis:
The judicial exception is not integrated into practical application. In particular, the claim recites the additional elements of:
“obtaining sensing data of a component to be detected”. This additional elements is recited at a high level of generality and amounts to extra-solution activity of gathering data i.e. pre-solution activity of gathering data for use in the claimed process.
“inputting the sensing data of a component to be detected into the trained neural network”. This additional element is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a generic computer component (component and neural network) (See MPEP 2106.05(f)).
“obtaining a density value based on an output from the trained neural network with respect to the input sensing data”. This additional elements is recited at a high level of generality and amounts to extra-solution activity of gathering data i.e. pre-solution activity of gathering data for use in the claimed process..
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
The additional element(s) of “obtaining sensing data of a component to be detected” and “obtaining a density value based on an output from the trained neural network with respect to the input sensing data” is/are recited at a high level of generality and amount(s) to extra-solution activity of receiving data i.e., pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
As discussed above, the additional element(s) of “inputting the sensing data of a component to be detected into the trained neural network” is/are recited at a high-level of generality such that it/they amount(s) to no more than mere instructions to apply the exception using generic computer components (See MPEP 2106.05(f)).
Accordingly, at Step 2B, the additional elements individually or in combination do not amount to significantly more than the judicial exception.
Regarding Claim 9:
Step 2A: Prong 1 analysis:
Claim 9 recites in part:
“determining a difference between the input sensing data and the reconstructed sensing data”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper. For example, this limitation encompasses identifying difference between input data and reconstructed data.
“identifying the component to be detected as an abnormal component, if the determined difference is above a threshold”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper. For example, this limitation encompasses identifying an anomaly based on a threshold value.
Accordingly, at Step 2A: Prong 1, the claim is directed to an abstract idea.
Step 2A: Prong 2 analysis:
The judicial exception is not integrated into practical application. In particular, the claim recites the additional elements of:
“obtaining sensing data of a component to be detected”. This additional elements is recited at a high level of generality and amounts to extra-solution activity of gathering data i.e. pre-solution activity of gathering data for use in the claimed process.
“inputting the sensing data of a component to be detected into the trained neural network”. This additional element is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a generic computer component (component and neural network) (See MPEP 2106.05(f)).
“obtaining reconstructed sensing data based on an output from the trained neural network with respect to the input sensing data”. This additional element is recited at a high level of generality such that the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished.
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
The additional element(s) of “obtaining sensing data of a component to be detected” and “obtaining reconstructed sensing data based on an output from the trained neural network with respect to the input sensing data” is/are recited at a high level of generality and amount(s) to extra-solution activity of receiving data i.e., pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
As discussed above, the additional element(s) of “inputting the sensing data of a component to be detected into the trained neural network” is/are recited at a high-level of generality such that it/they amount(s) to no more than mere instructions to apply the exception using generic computer components (See MPEP 2106.05(f)).
As discussed above, the additional element(s) of “obtaining reconstructed sensing data based on an output from the trained neural network with respect to the input sensing data” is/are recited at a high-level of generality such that the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished (See MPEP 2106.05(f)).
Accordingly, at Step 2B, the additional elements individually or in combination do not amount to significantly more than the judicial exception.
Regarding Claim 10:
Step 2A: Prong 1 analysis:
Claim 10 recites in part:
“clustering the sensing data based on feature maps generated by the trained neural network with respect to the input sensing data”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper. For example, this limitation encompasses identifying that resultant data can be formed into clusters based on its feature map.
“identifying the component to be detected as an abnormal component, if the sensing data is clustered outside a normal cluster”. As drafted and under its broadest reasonable interpretation, this limitation covers performance of the limitation in the mind (including an observation, evaluation, judgement, or opinion) or with the aid of pencil and paper. For example, this limitation encompasses identifying an anomaly in a cluster of data.
Accordingly, at Step 2A: Prong 1, the claim is directed to an abstract idea.
Step 2A: Prong 2 analysis:
The judicial exception is not integrated into practical application. In particular, the claim recites the additional elements of:
“obtaining sensing data of a component to be detected”. This additional elements is recited at a high level of generality and amounts to extra-solution activity of gathering data i.e. pre-solution activity of gathering data for use in the claimed process.
“inputting the sensing data of a component to be detected into the trained neural network”. This additional element is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a generic computer component (component and neural network) (See MPEP 2106.05(f)).
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
The additional element(s) of “obtaining sensing data of a component to be detected” is/are recited at a high level of generality and amount(s) to extra-solution activity of receiving data i.e., pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
As discussed above, the additional element(s) of “inputting the sensing data of a component to be detected into the trained neural network” is/are recited at a high-level of generality such that it/they amount(s) to no more than mere instructions to apply the exception using generic computer components (See MPEP 2106.05(f)).
Accordingly, at Step 2B, the additional elements individually or in combination do not amount to significantly more than the judicial exception.
Regarding Claim 15: Due to claim language similar to that of Claim 1, Claim 15 is rejected for the same reasons as presented above in the rejection of Claim 1, with the exception of the limitation(s) covered below.
Step 2A: Prong 2 analysis:
The judicial exception is not integrated into practical application. In particular, the claim recites the additional elements of:
“a memory”. This additional element is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a generic computer component (memory) (See MPEP 2106.05(f)).
“at least one processor coupled to the memory”. This additional element is recited at a high level of generality such that it amounts to no more than mere instructions to apply the exception using a generic computer component (processor) (See MPEP 2106.05(f)).
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
As discussed above, the additional element(s) of “a memory” and “at least one processor coupled to the memory” is/are recited at a high-level of generality such that it/they amount(s) to no more than mere instructions to apply the exception using generic computer components (See MPEP 2106.05(f)).
Accordingly, at Step 2B, the additional elements individually or in combination do not amount to significantly more than the judicial exception.
Regarding Claim 16: Due to claim language similar to that of Claim 8, Claim 16 is rejected for the same reasons as presented above in the rejection of Claim 8.
Regarding Claim 17: Due to claim language similar to that of Claim 9, Claim 17 is rejected for the same reasons as presented above in the rejection of Claim 9.
Regarding Claim 18: Due to claim language similar to that of Claim 10, Claim 14 is rejected for the same reasons as presented above in the rejection of Claim 10.
Regarding Claim 20:
Step 2A: Prong 2 analysis:
The judicial exception is not integrated into practical application. In particular, the claim recites the additional elements of:
“wherein the task is generating an image or producing a discriminative model”. This limitation merely indicates a field of use or technological environment in which the judicial exception is performed (machine learning model output types) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h).
Accordingly at Step 2A: Prong 2, the additional elements individually or in combination do not integrate the judicial exception into a practical application.
Step 2B analysis:
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception.
The additional element(s) of “wherein the task is generating an image or producing a discriminative model” is/are directed to particular field(s) of use (machine learning model output types) (MPEP 2106.05(h)) and therefore do not provide significantly more than the abstract idea, and thus the claim is subject-matter ineligible.
Accordingly, at Step 2B, the additional elements individually or in combination do not amount to significantly more than the judicial exception.
Regarding Claim 21: Due to claim language similar to that of Claim 20, Claim 21 is rejected for the same reasons as presented above in the rejection of Claim 20.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20180150728 A1 – machine learning systems and methods which are robust against label noise in training datasets in at least some circumstances
US 20130325782 A1 – a latent variable model estimation apparatus capable of implementing the model selection at high speed even if the number of model candidates increases exponentially as the latent state number and the kind of the observation probability increase
US 20180082172 A1 – A mechanism for compiling a generative description of an inference task into a neural network
US 20160300134 A1 – Systems and methods are disclosed for operating a Restricted Boltzmann Machine (RBM) by determining a corrected energy function of high-order semi-RBMs (hs-RBMs) without self-interaction
US 20190108396 A1 – A method for identifying and tracking objects
US 11636309 B2 – Systems and methods for modeling complex probability distributions
US 20200191943 A1 – Methods, apparatus and systems for wireless object tracking
Bao, F., Xu, K., Li, C., Hong, L., Zhu, J., & Zhang, B. (2021). Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models. arXiv [Cs.LG].– new estimates of the score function and its gradient with respect to the model parameters in a general energy-based latent variable model
Bao, F., Li, C., Xu, K., Su, H., Zhu, J., & Zhang, B. (2020). Bi-level Score Matching for Learning Energy-based Latent Variable Models. arXiv [Cs.LG]. Retrieved from– a bi-level score matching (BiSM) method to learn EBLVMs with general structures by reformulating SM as a bilevel optimization problem
Fischer, A., Igel, C. (2012). An Introduction to Restricted Boltzmann Machines. In: Alvarez, L., Mejail, M., Gomez, L., Jacobo, J. (eds) Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. CIARP 2012. Lecture Notes in Computer Science, vol 7441. Springer, Berlin, Heidelberg.– This tutorial introduces RBMs as undirected graphical models.
Oh, S., Baggag, A., & Nha, H. (2020). Entropy, Free Energy, and Work of Restricted Boltzmann Machines. Entropy, 22(5), 538. doi:10.3390/e22050538 – we analyze the training process of the restricted Boltzmann machine in the context of statistical physics
V. A. Shim, K. C. Tan and C. Y. Cheong, "An Energy-Based Sampling Technique for Multi-Objective Restricted Boltzmann Machine," in IEEE Transactions on Evolutionary Computation, vol. 17, no. 6, pp. 767-785, Dec. 2013, doi: 10.1109/TEVC.2013.2241768. – This paper examines the sampling techniques of a restricted Boltzmann machine-based multi-objective (MO) estimation of distribution algorithm (REDA).
Mullachery, V., Khera, A., & Husain, A. (2018). Bayesian Neural Networks. arXiv [Cs.LG]. – This paper describes, and discusses Bayesian Neural Network (BNN). The paper showcases a few different applications of them for classification and regression problems.
Saremi, S., Mehrjou, A., Schölkopf, B., & Hyvärinen, A. (2018). Deep Energy Estimator Networks. arXiv [Stat.ML]. – Bayesian interpretation of the score function and the Parzen score matching, and construct a multilayer perceptron with a scalable objective for learning the energy (i.e. the unnormalized log-density), which is then optimized with stochastic gradient descent
THIS ACTION IS MADE FINAL. 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 COREY M SACKALOSKY whose telephone number is (703)756-1590. The examiner can normally be reached M-F 7:30am-3:30pm EST.
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, Omar Fernandez Rivas can be reached at (571) 272-2589. 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.
/COREY M SACKALOSKY/Examiner, Art Unit 2128
/OMAR F FERNANDEZ RIVAS/Supervisory Patent Examiner, Art Unit 2128
Prosecution Insights