DETAILED ACTION
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 .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-7 and 9-12 are rejected under 35 U.S.C. 103 as being unpatentable over Zhao et al., RoboGrammar: Graph Grammar for Terrain-Optimized Robot Design (Zhao) in view of Gal et al., Dropout as a Bayesian Approximation (Gal) in further view of Mehta et al., US 2022/0284286 Al (Mehta).
Claim 1.
Zhao teaches A computer implemented method for automated support of a design of a technical system, comprising the following operations, wherein the operations are performed by modules, and wherein the modules are software modules executed by one or more processors and/or hardware modules: receiving, by a machine learning model, a current partial design of a technical system and a candidate component for a next design step of designing the technical system; (Zhao Pg 188:8 Section 6.1) “The input to the function is a graph representing a partial robot design, where some nodes correspond to non-terminal symbols. The partial design may be expanded into one of many complete designs which have only terminal symbols. The function V aims to outputs the highest achievable performance across all of these complete designs. (188:1 Section 1) “The goal of the system is to take a set of user specified primitive components and generate an optimal robot structure. . .”
Zhao does not explicitly teach, but Gal teaches computing, by the machine learning model; a probability distribution, which is a probability distribution over changes of a design key performance indicator if the candidate component is added to the current partial design, with the design key performance indicator describing a property of the technical system; and/or a predicted impact value predicting an absolute value of the design key performance indicator or a change of the design key performance indicator if the candidate component is added to the current partial design; From the above list of alternatives the Examiner is selecting probability distribution over key changes. (Gal Section 5.3 “Similarly to (Hern´andez-Lobato & Adams, 2015) we use Bayesian optimization over validation log-likelihood to find optimal, and set the prior length-scale to 10-2 for most datasets based on the range of the data.”)
Zhao and Gal are analogous to the claimed invention because they are from the same field of endeavor of optimizing design of a technical system.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Zhao and Gal before him or her, to modify the partial designs of Zhao with the probability distribution of Gal as a way to mitigate the uncertainty of deep learning as Gal (abstract) suggests.
Modified Zhao with Gal does not explicitly teach, but Mehta teaches and outputting, by the machine learning model and/or a user interface, the probability distribution and/or the predicted impact value. (Mehta 0017) “one or more sequences of complementary items generated by the second artificial intelligence module are output on the screen of the user interface. . .”
Zhao, Gal, and Mehta are analogous to the claimed invention because they are from the same field of endeavor of optimizing design of a technical system.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Zhao, Gal, and Mehta before him or her, to modify the partial designs of Zhao with the probability distribution of Gal with the user interface of Mehta to reduce the amount of time, and effort needed by a user. (Mehta 0003)
Claim 2.
Modified Zhao teaches The method according to claim 1, wherein the current partial design is a graph-based representation; (Zhao pg. 188:8 section 6.1 Evaluation phase) “The input to the function is a graph representing a partial robot design.”
and the machine learning model is graph-based. (Zhao pg. 188:8 section 6.1 Evaluation phase) “In practice, we take a deep-learning-based approach and use graph neural networks to create our learnable heuristic.”
Claim 3.
Modified Zhao teaches The method according to claim 1, wherein the machine learning model contains at least one graph neural network; at least one graph convolutional neural network; and/or at least one Bayesian neural network. From the above list of alternatives, the Examiner is selecting “graph neural network.” (Zhao pg. 188:8 section 6.1 Evaluation phase) “In practice, we take a deep-learning-base approach and use graph neural networks to create our learnable heuristic.”
Claim 4.
Modified Zhao teaches The method according to claim 1, wherein the machine learning model computes individual probability distributions and/or predicted impact values for several next component options; and wherein the user interface outputs the individual probability distribution and/or predicted impact value for each next component option. From the above list of alternatives, the Examiner is selecting “predicted impact values for several next component options.” (Zhao pg. 188:8 section 6.1.1 Evaluation phase) “K possible candidate robot designs are generated and one of them is selected for evaluation. {Examiners note: These K designs constitute the next component options. After evaluation the average reward is stored and associated with its candidate.}
Claim 5.
Modified Zhao teaches and completing, by a processor, the current partial design with the selected next component option, thereby producing a completed design. (Zhao pg. 188:10 Section 6.2) “are then randomly selected and applied until a complete design is obtained. The complete design is evaluated in simulation using MPC, in a way similar to Graph Heuristic Search.” (pg. 188:8 Design phase) “production rules of the grammar are iteratively applied to the partial robot design until it contains only terminal symbols”
Modified Zhao with Mehta teaches The method according to claim 4, with the additional operations of detecting, by the user interface, a user interaction selecting one of the next component options; (Mehta [0016-0017]) “In a further possible embodiment of the recommendation engine according to the first aspect of the present invention, the items are selected by a user via a user interface having a screen adapted to output available items to the user. . . one or more sequences of complementary items generated by the second artificial intelligence module are output on the screen of the user interface for selection of a next item from one of the sequences of complementary items or for selection of one or more items (not necessarily appearing one after the other) from one of the sequences of complementary items or for selection of a whole sequence of complementary items by the user.
Claim 6.
Modified Zhao with Mehta teaches The method according to claim 5, with the additional operation of: automatically manufacturing, by an automated plant, the completed design, thereby producing the technical system. (Zhao pg. 188:15 section 11) “Importantly, the emergent designs are observably physically fabricable and there is significant potential for the designs to be translated to real-world scenarios and environments.”
Claim 7.
Modified Zhao teaches The method according to claim 1, with the initial operation of training, by a differ training module, the machine learning model with training examples, with each training example comprising: a partial design; a candidate next component; and a best value for the design key performance indicator computed by a simulation environment for a completed design, wherein the completed design is reachable and feasible when adding the candidate next component to the partial design. (Zhao pg. 188:9 Learning phase) “The heuristic is trained using the data stored in the lookup table. . . . “We consider the Vˆ of a design to be the best average reward over all evaluations of the design. If this average reward is the best seen so far, the candidate is stored as the current best design, along with Vˆ. (pg. 188:9 section 6.1.2 “We convert robot design graphs into inputs for the GNN model as follows.” {Examiners note: Each Training example consists of a partial design, the candidate component applied via grammer rule, and the best kpi value obtained from simulation.}
Claim 9.
Modified Zhao teaches A system for automated support of a design of a technical system, comprising: a machine learning model, configured for receiving a current partial design of a technical system and a candidate component for a next design step of designing the technical system; (Zhao Pg 188:8 Section 6.1) “The input to the function is a graph representing a partial robot design, where some nodes correspond to non-terminal symbols. The partial design may be expanded into one of many complete designs which have only terminal symbols. The function V aims to outputs the highest achievable performance across all of these complete designs. (188:1 Section 1) “The goal of the system is to take a set of user specified primitive components and generate an optimal robot structure. . .”
Zhao does not explicitly teach, but Gal teaches computing, by the machine learning model; a probability distribution, which is a probability distribution over changes of a design key performance indicator if the candidate component is added to the current partial design, with the design key performance indicator describing a property of the technical system; and/or a predicted impact value predicting an absolute value of the design key performance indicator or a change of the design key performance indicator if the candidate component is added to the current partial design; From the above list of alternatives the Examiner is selecting probability distribution over key changes. (Gal Section 5.3 “Similarly to (Hern´andez-Lobato & Adams, 2015) we use Bayesian optimization over validation log-likelihood to find optimal, and set the prior length-scale to 10-2 for most datasets based on the range of the data.”)
Zhao and Gal are analogous to the claimed invention because they are from the same field of endeavor of optimizing design of a technical system.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Zhao and Gal before him or her, to modify the partial designs of Zhao with the probability distribution of Gal as a way to mitigate the uncertainty of deep learning as Gal (abstract) suggests.
a user interface with a display, configured for outputting the probability distribution and/or the predicted impact value. (Mehta 0017) “one or more sequences of complementary items generated by the second artificial intelligence module are output on the screen of the user interface. . .”
Zhao, Gal, and Mehta are analogous to the claimed invention because they are from the same field of endeavor of optimizing design of a technical system.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Zhao, Gal, and Mehta before him or her, to modify the partial designs of Zhao with the probability distribution of Gal with the user interface of Mehta to reduce the amount of time, and effort needed by a user. (Mehta 0003)
Claim 10.
Claim 10 is rejected as being substantially similar to claim 1, albeit for a computer program.
Claim 11.
Claim 11 is rejected as being substantially similar to claim 1, albeit for a computer program product.
Claim 12.
Claim 12 is rejected as being substantially similar to claim 1, albeit for a computer readable storage media.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Zhao et al., RoboGrammar: Graph Grammar for Terrain-Optimized Robot Design (Zhao) in view of Gal et al., Dropout as a Bayesian Approximation (Gal) in further view of Mehta et al., US 2022/0284286 Al (Mehta) in further view of Mendonca et al., SAT-based Analysis of Feature Models is Easy (Mendonca).
Claim 8.
Modified Zhao teaches the training examples by taking partial designs and candidate next components as input and sampling with feasibility constraints in order to produce expanded designs, with the expanded designs consisting of samples of a design space that are each reachable and feasible given a set of component compatibility constraints when combining a partial design with one of the candidate next components; iterating the generating operation, using the expanded designs as partial designs, until completed designs are reached; (Zhao Pg 188:8 Section 6.1) “The input to the function is a graph representing a partial robot design, where some nodes correspond to non-terminal symbols. The partial design may be expanded into one of many complete designs which have only terminal symbols. The function V aims to outputs the highest achievable performance across all of these complete designs. (188:1 Section 1) “The goal of the system is to take a set of user specified primitive components and generate an optimal robot structure. . .” {EXAMINERS NOTE: Designs and rule based expansion (adding components).}
and simulating, by the simulation environment, each completed design in order to compute its value for the design key performance indicator. (Zhao pg. 188:8 section 6.1.1 Evaluation phase) “If this average reward is the best seen so far, the candidate is stored as the current best design, along with Vˆ . Regardless, the candidate robot design and its corresponding Vˆ are stored (or updated) in a lookup table, and the vˆ label of all of that design’s partial design ancestors are updated to be the maximum of their current value and the candidate robot’s average reward.”
Modified Zhao does not explicitly teach, but Mendonca teaches The method according to claim 7, wherein the training examples are initially prepared with the following operations: generating, by a processor executing a SAT solver, (Mendonca Abstract) “Automated analyses of feature models, such as consistency checking and interactive or offline product selection, often rely on translating models to propositional logic and using satisfiability (SAT) solvers.”
Zhao, Gal, Mehta, and Mendonca are analogous to the claimed invention because they are from the same field of endeavor of optimizing design of a technical system.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Zhao and Gal before him or her, to modify the partial designs of Zhao with the probability distribution of Gal with the SAT solver of Mendonca to make it more efficient to translate model to propositional logic. (Medonca Abstract)
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN DAVID HAGLER whose telephone number is (703)756-1339. The examiner can normally be reached Monday - Friday 10am- 6pm.
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, Rehana Perveen can be reached at 5712723676. 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.
/JOHN DAVID HAGLER/Examiner, Art Unit 2189
/REHANA PERVEEN/Supervisory Patent Examiner, Art Unit 2189