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 .
Detailed Action
This action is in response to the claims filed 6/15/2023:
Claims 1 – 20 are pending.
Claims 1, 14, and 20 are independent.
Drawings
The drawings are objected to because FIG. 3c is a low quality scan containing illegible elements. Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Objections
Claims 1, 14, and 20 are objected to because of the following informalities: "having plurality of features" should read "having a plurality of features". Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 9 and 18 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Regarding claims 9 and 18, "a uGLAD optimization algorithm" is indefinite. uGLAD is an undefined acronym and also an undefined algorithm that would not have been known to one of ordinary skill in the art before the effective filing date of the claimed invention. In view of the spec a "uGLAD optimization algorithm" is interpreted as any algorithm that ([¶0050] "is a deep learning model that can recover sparse graphs in an unsupervised manner. It builds upon and extends the GLAD model which does sparse graph recovery by applying deep unfolding technique on the Alternating Minimization updates under supervision.").
Claim Rejections - 35 USC § 101
101 Rejection
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 1-20 are rejected under 35 USC § 101 because the claimed invention is directed to non-statutory subject matter.
Regarding Claim 1: Claim 1 is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 1 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: Claim 1 under its broadest reasonable interpretation is a series of mental processes and mathematical calculations and relationships. For example, but for the generic computer components language, the above limitations in the context of this claim encompass machine learning processing, including the following:
generating preprocessed input data, wherein generating the preprocessed input data includes one or more of performing data normalization and (observation, evaluation, and judgement)
calculating a covariance matrix (mathematical calculation),
assessing data quality of the preprocessed input data, wherein assessing data quality includes one or more of collecting summary statistics and performing missingness analysis (observation, evaluation, and judgement)
generating a domain structure from the preprocessed input data, the domain structure including representations of functional dependencies between features of the plurality of data samples (observation, evaluation, and judgement)
determine inference results including conditional distribution and maximum a posteriori (MAP) values for one or more variables of interest given evidence (observation, evaluation, and judgement)
generating a dependency function between two features represented in the domain structure based on the PGM (observation, evaluation, and judgement)
Therefore, claim 1 recites an abstract idea which is a judicial exception.
Step 2A Prong Two Analysis: Claim 1 recites additional elements “applying the PGM to observed or hypothetical evidence in a form of specific values assigned to a subset of features to”. However, these additional features are computer components recited at a high-level of generality, such that they amount to no more than mere instructions to apply the judicial exception using a generic computer component. An additional element that merely recites the words “apply it” (or an equivalent) with the judicial exception, or merely includes instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea, does not integrate the judicial exception into a practical application (See MPEP 2106.05(f)). Claim 1 also recites additional elements “obtaining input data including a plurality of data samples, each of the plurality of data samples having plurality of features;”, “recovering a probabilistic graphical model (PGM) trained to fit a probabilistic density function over the plurality of features based on the preprocessed input data and the domain structure”, and “presenting an output via a display device based on one or more of the summary statistics, the missingness analysis, the domain structure, the PGM, the inference results, and the dependency function” which amounts to gathering and outputting data which is insignificant extra-solution activity (See MPEP 2106.05(g)). Therefore, claim 1 is directed to a judicial exception.
Step 2B Analysis: Claim 1 does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the lack of integration of the abstract idea into a practical application, the additional elements recited in claim 1 amount to no more than mere instructions to apply the judicial exception using a generic computer component and insignificant extra-solution activity. The gathering and outputting of data is considered well-understood, routine, and conventional in the art (See MPEP 2106.05(d)(II)(i) and MPEP 2106.05(d)(II)(iv).
For the reasons above, claim 1 is rejected as being directed to non-patentable subject matter under §101. This rejection applies equally to independent claims 14 and 20, which recite a system and a non-transitory computer readable medium, respectively, as well as to dependent claims 2-13, and 15-19.
The additional elements in claim 14 (“at least one processor; memory in electronic communication with the at least one processor; and instructions stored in the memory, the instructions being executable by the at least one processor to;”) are seen as mere instructions to apply the judicial exception using generic computer components.
The additional elements in claim 20 (“A non-transitory computer readable medium storing instructions thereon that, when executed by at least one processor, cause a computing device to:”) are seen as mere instructions to apply the judicial exception using generic computer components.
The additional limitations of the dependent claims are addressed briefly below:
Dependent claim 2 recites additional observation, evaluation, and judgement “the plurality of features include multiple datatypes including two or more of a continuous numeric type, discrete numeric type, nominal categorical type, or ordinal categorical type”
Dependent claim 3 recites additional observation, evaluation, and judgement “assessing data quality includes performing missingness analysis for each feature of the plurality of features to determine a missingness rate, a missingness rate over time, a missingness type, and a missingness dependence on other features”
Dependent claim 4 recites additional observation, evaluation, and judgement “the missingness type is one or more of a missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR)”
Dependent claim 5 recites additional observation, evaluation, and judgement “calculating the covariance matrix includes calculating covariance between two or more features of the plurality of features”
Dependent claims 6 and 15 recite additional observation, evaluation, and judgement “the domain structure is a conditional independence (CI) graph”
Dependent claims 7 and 16 recite additional observation, evaluation, and judgement “wherein the CI graph is modeled using at least one of a regression-based approach with graph sparsity constraints, a partial correlation estimation approach, a graphical lasso approach, or a Markov networks approach”
Dependent claim 8 and 17 recite additional instructions to apply the judicial exception using generic computer components “the PGM is a neural graphical model (NGM)” (the instant specification does not define or limit an NGM in a way that would make it unreasonable to interpret it as a standard PGM which is interpreted as a generic computer component)
Dependent claims 9 and 18 recite additional observation, evaluation, and judgement “wherein recovering the NGM includes modeling the CI graph obtained using a uGLAD optimization algorithm” (the instant specification does not define a uGLAD algorithm in a way that would make it unreasonable to interpret it as a mental process, nor would a uGLAD algorithm be readily known to one of ordinary skill in the art as Applicant appears to be acting as their own lexicographer)
Dependent claims 10 and 19 recite additional observation, evaluation, and judgement “wherein generating the domain structure and recovering the NGM are performed using a neural graph revealer (NGR) optimization algorithm” (the instant specification does not define an NGR algorithm in a way that would make it unreasonable to interpret it as a mental process, nor would a NGR algorithm be readily known to one of ordinary skill in the art as Applicant appears to be acting as their own lexicographer))
Dependent claim 11 recites additional instructions to apply the judicial exception using generic computer components “wherein using the NGR optimization algorithm includes applying a training algorithm to a randomly initialized neural network based architecture (eg. a fully connected multilayer perceptron) using the preprocessed input data to generate an optimized regression model that indicates functional dependencies between different features of the preprocessed input data” (a fully connected multilayer perceptron is a regression model that is expected to be trained by definition)
Dependent claim 12 recites additional mathematical calculations and relationships “applying the PGM to the observed or hypothetical evidence includes computing one or more of maximum a posteriori (MAP) values and conditional probability distributions”
Dependent claim 13 recites additional insignificant extra-solution activity of gathering and outputting data (See MPEP 2106.05(g)) “receiving an interaction input from a user in connection with the observed or hypothetical evidence, or in connection with dependency function computation, wherein presenting the output via the display device is further based on the received interaction input” which is well-understood, routine, and conventional in the art (See MPEP 2106.05(d)(II)).
Therefore, when considering the elements separately and in combination, they do not add significantly more to the inventive concept. Accordingly, claims 1-20 are rejected under 35 U.S.C. § 101.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1, 2, 5, 12, 13, 14, and 20 are rejected under U.S.C. §102(a)(1) as being anticipated by Eck (US20190384235A1).
Regarding claim 1, Eck teaches In a computing environment including one or more server devices hosting services thereon, ([¶0079] "a “server” includes a physical data processing system (for example, system 212 as shown in FIG. 2) running a server program. It will be understood that such a physical server may or may not include a display and keyboard.")
a method for generating and presenting insights on collections of data samples, the method comprising: ([Abstract] "An initially empty inference model is extended with the set of variables, to obtain an extended model. A request to observe a given one of the set of variables at a given timestamp is obtained. Responsive thereto, time series data for the set of registered variables is retrieved. The extended model is run with the retrieved data to obtain an estimate of the given one of the variables at the given timestamp.")
obtaining input data including a plurality of data samples, each of the plurality of data samples having plurality of features;([¶0004] "a data store storing at least one time series of sensor data" [¶0031] "The system retrieves time series data for all variables in the inference model." time series contains many samples, each explicitly corresponding to variables (features))
generating preprocessed input data, wherein generating the preprocessed input data includes one or more of performing data normalization and calculating a covariance matrix; ([¶0029] "the user registers known analytic relations between variables. Examples of analytic relations could be functional relations, parametric joint densities (mean/covariance matrix), and the like" [¶0055] " in the joint density, there is a Gaussian distribution, the parameters would be a mean and a covariance matrix.")
assessing data quality of the preprocessed input data, wherein assessing data quality includes one or more of collecting summary statistics and performing missingness analysis; ([¶0034] "When User A requests data for active power, the system runs an inference on the factor graph model and returns the estimate of that variable based on all available observations and factor relations. The inference takes care of complementing for missing data in specific variables at certain timestamps. The system returns “unobservable active power at time X” if too many relevant data points are missing.")
generating a domain structure from the preprocessed input data, the domain structure including representations of functional dependencies between features of the plurality of data samples; ([¶0029] "a semantic store 103 (system entities, variables, relations) and a mapping between semantic data and time series data (dashed lines) […] the system navigates the semantic store to identify new relationships and associates a parametric relation (e.g. a joint/conditional density) to each relation. The system then learns the parameters of the new analytic relations by extracting historical data from the time series data store." The semantic store provides a graph-like domain structure of variables and relations. The disclosed analytic relations include functional relations and probabilistic relations that are attached, i.e., representations of dependencies among features/variables)
recovering a probabilistic graphical model (PGM) trained to fit a probabilistic density function over the plurality of features based on the preprocessed input data and the domain structure;([¶0054] "the inference model includes a probabilistic graphical model, and the learning of the parameters of the new relationships by extracting the historical data from the at least one time series of sensor data includes using a maximum likelihood technique." [¶0051] "the parametric relation includes at least one of a joint probability density and a conditional probability density." Eck explicitly uses a PGM as the inference model, explicitly associates joint/conditional probability densities to relations, and explicitly learns parameters from historical data via ML. This is interpreted as synonymous with "recovering" a PGM "trained to fit" probabilistic densities, guided by the sematic relations/domain structure.)
applying the PGM to observed or hypothetical evidence in a form of specific values assigned to a subset of features to determine inference results including conditional distribution ([¶0031] "the user requests to observe a variable at a given timestamp. The system retrieves time series data for all variables in the inference model" [¶0054] "the inference model includes a probabilistic graphical model, and the learning of the parameters of the new relationships by extracting the historical data from the at least one time series of sensor data includes using a maximum likelihood technique." [¶0051] "the parametric relation includes at least one of a joint probability density and a conditional probability density." [¶0057] "If a certain value of a given variable is assumed, the system allows one to determine the effect of same on the other variables")
and maximum a posteriori (MAP) values for one or more variables of interest given evidence;([¶0031] "The system then runs the inference model and returns the “optimal” estimate of the requested variable given the timeseries data and the model relationships, where the optimality criterion could be, for example, maximum likelihood or maximum a posteriori")
generating a dependency function between two features represented in the domain structure based on the PGM; and([¶0044] "a query about the sensitivity of entities in the system is received, and the inference model is run so as to derive a quantitative measure of the sensitivity of the semantic entity with respect to all other semantic entities included in the inference model." [¶0054] "the inference model includes a probabilistic graphical model" sensitivity measure interpreted as dependency function explicitly based on features of the PGM)
presenting an output via a display device based on one or more of the summary statistics, the missingness analysis, the domain structure, the PGM, the inference results, and the dependency function.([¶0073] "an implementation might employ, for example, a processor 202, a memory 204, and an input/output interface formed, for example, by a display 206 […] the phrase “input/output interface” as used herein, is intended to include, for example, one or more mechanisms for inputting data to the processing unit (for example, mouse), and one or more mechanisms for providing results" [¶0034] " When User A requests data for active power, the system runs an inference on the factor graph model and returns the estimate of that variable based on all available observations and factor relations. The inference takes care of complementing for missing data in specific variables at certain timestamps. The system returns “unobservable active power at time X” if too many relevant data points are missing.").
Regarding claim 2, Eck teaches The method of claim 1, wherein the plurality of features include multiple datatypes including two or more of a continuous numeric type, discrete numeric type, nominal categorical type, or ordinal categorical type.(Eck [¶0004] "a request to observe a given one of the set of one or more variables of interest at a given timestamp" [¶0034] "User A (an electrical grid operator) registers variables of interest for monitoring: (i) Active Power 303 at a substation 301, (ii) Voltage at a feeder head 305, and, optionally, any other desired appropriate quantities (e.g., (iii) Frequency at a coupling point, not shown)." The time series explicitly corresponds both to continuous nominal data such as voltage at a discrete numeric timestamp).
Regarding claim 5, Eck teaches The method of claim 1, wherein calculating the covariance matrix includes calculating covariance between two or more features of the plurality of features.(Eck [¶0029] "the user registers known analytic relations between variables. Examples of analytic relations could be functional relations, parametric joint densities (mean/covariance matrix), and the like" [¶0055] " in the joint density, there is a Gaussian distribution, the parameters would be a mean and a covariance matrix.").
Regarding claim 12, Eck teaches The method of claim 1, wherein applying the PGM to the observed or hypothetical evidence includes computing one or more of maximum a posteriori (MAP) values and conditional probability distributions.(Eck [¶0031] "The system then runs the inference model and returns the “optimal” estimate of the requested variable given the timeseries data and the model relationships, where the optimality criterion could be, for example, maximum likelihood or maximum a posteriori").
Regarding claim 13, Eck teaches The method of claim 1, further comprising receiving an interaction input from a user in connection with the observed or hypothetical evidence, or in connection with dependency function computation, wherein presenting the output via the display device is further based on the received interaction input.(Eck [¶0073] "an implementation might employ, for example, a processor 202, a memory 204, and an input/output interface formed, for example, by a display 206 […] the phrase “input/output interface” as used herein, is intended to include, for example, one or more mechanisms for inputting data to the processing unit (for example, mouse), and one or more mechanisms for providing results" [¶0034] " When User A requests data for active power, the system runs an inference on the factor graph model and returns the estimate of that variable based on all available observations and factor relations. The inference takes care of complementing for missing data in specific variables at certain timestamps. The system returns “unobservable active power at time X” if too many relevant data points are missing.").
Regarding claim 14, claim 14 is directed towards a system for performing the method of claim 1. Therefore, the rejection applied to claim 1 also applies to claim 14. Claim 14 also recites additional elements at least one processor;
memory in electronic communication with the at least one processor; and
instructions stored in the memory, the instructions being executable by the at least one processor to; (Eck [¶0008] "one or more embodiments of the invention or elements thereof can be implemented in the form of a system (or apparatus) including a memory, and at least one processor that is coupled to the memory and operative to perform exemplary method steps. Yet further, in another aspect, one or more embodiments of the invention or elements thereof can be implemented in the form of means for carrying out one or more of the method steps described herein; the means can include (i) hardware module(s), (ii) software module(s) stored in a computer readable storage medium (or multiple such media) and implemented on a hardware processor, or (iii) a combination of (i) and (ii); any of (i)-(iii) implement the specific techniques set forth herein.").
Regarding claim 20, claim 20 is directed towards a non-transitory computer readable medium storing instructions for performing the method of claim 1. Therefore, the rejection applied to claim 1 also applies to claim 20. Claim 20 also recites additional elements A non-transitory computer readable medium storing instructions thereon that, when executed by at least one processor, cause a computing device to:(Eck [¶0079] "a “server” includes a physical data processing system (for example, system 212 as shown in FIG. 2) running a server program. It will be understood that such a physical server may or may not include a display and keyboard.").
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 3 and 4 are rejected under U.S.C. §103 as being unpatentable over the combination of Eck and Enders (“Applied Missing Data Analysis”, 2010).
Regarding claim 3, Eck teaches The method of claim 1.
However, Eck doesn't explicitly teach wherein assessing data quality includes performing missingness analysis for each feature of the plurality of features to determine a missingness rate, a missingness rate over time, a missingness type, and a missingness dependence on other features.
Enders, in the same field of endeavor, teaches assessing data quality includes performing missingness analysis for each feature of the plurality of features to determine a missingness rate, a missingness rate over time, a missingness type, and a missingness dependence on other features. ([p. 27] "closely approximates the missing data rate" [p. 52] "if a participant drops out after the fifth week of an 8-week study, his week five score fills in the remaining waves of data" [p. 303] "the probability of missingness at wave t depends on the outcome variable at time t and the outcome variable from the previous data collection wave" Missing data rate interpreted as missingness rate. probability of missingness at time t interpreted as missingness rate over time.).
Eck as well as Enders are directed towards data analysis. Therefore, Eck as well as Enders are reasonably pertinent analogous art. It would have been obvious before the effective filing date of the claimed invention to combine the teachings of Eck with the teachings of Enders. Given that real-world time series sensor data routinely has gaps/dropout (as is explicitly acknowledged in Eck), it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Ender' missingness diagnostics as part of Eck's data cleaning stage ([p. 308] "combining missing data patterns can improve the reliability of the pattern-specific estimates").
Regarding claim 4, the combination of Eck, and Enders teaches The method of claim 3, wherein the missingness type is one or more of a missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR).(Enders [p. 303] "to estimate several of these MNAR models" [p. 304] "3 suggests that the data are MAR (i.e., dropout at time t is related to the scores from a previous assessment). If the entire set of logistic regression coefficients is nonsignificant, this suggests that the data are MCAR (i.e., dropout is unrelated to variables in the model). Hypo thetically, it is possible to test the MAR mechanism by comparing the fi t of the model in Figure 10.3 to that of a nested model that constrains the MNAR coefficients").
Claims 6-10 and 15-19 are rejected under U.S.C. §103 as being unpatentable over the combination of Eck and Pu (“Learning to Learn Graph Topologies”, 2021).
Regarding claim 6, Eck teaches The method of claim 1.
However, Eck doesn't explicitly teach, wherein the domain structure is a conditional independence (CI) graph.
Pu, in the same field of endeavor, teaches the domain structure is a conditional independence (CI) graph.([p. 3] "The `1 norm on edge weights enforces the learned conditional independence graph to be sparse").
Eck as well as Pu are directed towards machine learning for entity relationships. Therefore, Eck as well as Pu are reasonably pertinent analogous art. It would have been obvious before the effective filing date of the claimed invention to combine the teachings of Eck with the teachings of Pu. A person of ordinary skill in the art would be motivated to plug a learned dependence graph in Eck's inference model pipeline to improve scalability and accuracy and to reduce manual modeling effort (especially when Eck's semantic store already track relations between variables. Pu provides as additional motivation for combination ([p. 2] “the proposed method improves the accuracy of graph learning by 80% compared to traditional iterative solvers”). This motivation for combination also applies to the remaining claims which depend on this combination.
Regarding claim 7, the combination of Eck and Pu teaches The method of claim 6, wherein the CI graph is modeled using at least one of a regression-based approach with graph sparsity constraints, a partial correlation estimation approach, a graphical lasso approach, or a Markov networks approach.(Pu [p. 3] "The l1 norm on edge weights enforces the learned conditional independence graph to be sparse" [p. 6] "GLAD [31] unrolls an alternating minimisation algorithm for solving an l1 regularised log-likelihood maximisation of the precision matrix" l1 norm on edge weights interpreted as explicit graph sparsity constraint. L1 regularised log-likelihood maximisation of the precision matrix interpreted as a graphical lasso approach.).
Regarding claim 8, the combination of Eck and Pu teaches The method of claim 7, wherein the PGM is a neural graphical model (NGM). (Eck [¶0027] "The analytics engine is configured to interpret analytics relations and run inference(s) on the inference model. Non-limiting examples of analytic relations include: deterministic functions, joint/conditional probability distributions, and the like. The machine learning modules are configured to learn new analytics relations." [¶0054] "the inference model includes a probabilistic graphical model" the PGM in Eck is interpreted as an NGM in view of the instant specification at [¶0030] " a “neural graphical model” (NGM) refers to a computer algorithm or model (e.g., a classification model, regression model, probabilistic graphical model, etc.) that can be tuned or trained based on training input to approximate unknown features or values".).
Regarding claim 9, the combination of Eck and Pu teaches The method of claim 8 wherein recovering the NGM includes modeling the CI graph obtained using a uGLAD optimization algorithm. (Pu [p. 3] "The l1 norm on edge weights enforces the learned conditional independence graph to be sparse" [p. 6] "GLAD [31] unrolls an alternating minimisation algorithm for solving an l1 regularised log-likelihood maximisation of the precision matrix" [p. 2] "we first unroll an iterative algorithm for solving the aforementioned regularised graph learning objective." [p. 2] "unrolling an iterative algorithm that is originally designed for graph learning problem introduces a sensible inductive bias that makes our model highly interpretable" uGLAD interpreted in view of the instant specification [¶0050] "One possible optimization algorithm is an uGLAD, which is a deep learning model that can recover sparse graphs in an unsupervised manner. It builds upon and extends the GLAD model which does sparse graph recovery by applying deep unfolding technique on the Alternating Minimization updates under supervision.").
Regarding claim 10, the combination of Eck and Pu teaches The method of claim 8, wherein generating the domain structure and recovering the NGM are performed using a neural graph revealer (NGR) optimization algorithm.(Pu [p. 2] "we propose a novel functional learning framework to learn a mapping from node observations to the underlying graph topology with desired structural property […] we propose a novel and highly interpretable neural networks to learn a data-to-graph mapping based on algorithmic unrolling" interpreted as NGR in view of the instant specification [¶0031] "a “neural graph revealer” (NGR) refers to a method for jointly recovering a graph and fitting a regression function for all features using a neural network at the same time. As such, an NGR combines the steps of domain structure recovery (which results in a graph indicating dependencies between functions) and creating an NGM capable of performing inference.".).
Regarding claims 15-19, claims 15-19 are directed towards a system for performing the methods of claims 6-10, respectively. Therefore, the rejections applied to claims 6-10 also apply to claims 15-19.
Claim 11 is rejected under U.S.C. §103 as being unpatentable over the combination of Eck and Pu and in further view of Barrachina (“THEORY AND IMPLEMENTATION OF COMPLEX-VALUED NEURAL NETWORKS”, 2023).
Regarding claim 11, the combination of Eck and Pu teaches The method of claim 10, wherein using the NGR optimization algorithm includes applying a training algorithm to a [randomly] initialized neural network based architecture (eg. a fully connected multilayer perceptron) using the preprocessed input data to generate an optimized regression model that indicates functional dependencies between different features of the preprocessed input data.(Pu [p. 2] "We train the model in an end-to-end fashion with pairs of data and graphs that share the same structural properties. Once trained, the model can be deployed to learn a graph topology that exhibits such structural properties." [p. 3] "we aim at learning a neural network F✓(·) that maps the data term y to a graph representation w that share the same topological properties and hence belong to the same graph family G. With training pairs {(yi,wi)}" Pu is explicitly training a neural network on node data to output a graph representation that encodes relationships by learning a parametric function minimizing a likelihood-based objective to estimate said relationships among variables which is interpreted as constituting an optimized regression model).
However, the combination of Eck and Pu doesn't explicitly teach random neural network initialization.
Barrachina, in the same field of endeavor, teaches random neural network initialization([p. 20 §7] "we are to blindly apply any well-known random initialization algorithm to both real and imaginary parts of each trainable parameter independently" [p. 22] "It will therefore suffice to choose any random initialization, such as, for example, a Rayleigh distribution").
The combination of Eck and Pu as well as Barrachina are directed towards neural networks. Therefore, the combination of Eck and Pu as well as Barrachina are reasonably pertinent analogous art. It would have been obvious before the effective filing date of the claimed invention to combine the teachings of the combination of Eck and Pu with the teachings of Barrachina by randomly initializing the neural networks. Barrachina explicitly teaches that random neural network parameter initialization is well-known in the art ([p. 20 §7] "we are to blindly apply any well-known random initialization algorithm to both real and imaginary parts of each trainable parameter independently" [p. 22] "It will therefore suffice to choose any random initialization, such as, for example, a Rayleigh distribution").
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Shental (US 20100074342 A1) is directed towards noisy data analysis using covariance matrix, probability graphs, and covariance matrices .
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SIDNEY VINCENT BOSTWICK whose telephone number is (571)272-4720. The examiner can normally be reached M-F 7:30am-5:00pm 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, Miranda Huang can be reached on (571)270-7092. 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.
/SIDNEY VINCENT BOSTWICK/Examiner, Art Unit 2124
Prosecution Insights