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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 03/12/2026, 11/24/2025, and 09/22/2025 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Objections
Claims 9-13 and 15-20 are objected to because of the following informalities:
Regarding claim 9, line 1, “One or more non-transitory computer-readable media” should be “A non-transitory computer-readable media”;
Regarding claims 10-13, line 1, “The one or more non-transitory computer-readable media” should be “The non-transitory computer-readable media”;
Regarding claims 15-20, line 1, “The one or more computing systems” should be “The computing system”;
Regarding claim 20, line 2, “the RPA robot” should be “the automation”.
Appropriate correction is required.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1-20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1-20 of U.S. Patent No. 12423609.
Although conflicting claims are not identical, they are not patentably distinct from each other because removing inherent and/or unnecessary limitation(s)/step(s) or adding an element and its function would be within the level of one of ordinary skill in the art. It is well settled that the adding or deleting of an element and its function(s) in the claim of the present application are an obvious expedient if the remaining elements perform the same function as before. In re Karlson, 136 USPQ 184 (CCPA 1963). Also note Ex parte Rainu, 168 USPQ 375 (Bd. App. 1969). Omission of a referenced element or step whose function is not needed would be obvious to one of ordinary skill in the art. Examiner further notes wherein although the claims are not identical (broader in scope and in a system or machine form), they are commensurate in scope to the claim limitations provided in the issued U.S. Patent, and likewise would anticipate the currently provided claim limitations.
Table 1: Comparison of claims in Instant Application 19307558 vs. U.S. Patent No. 12423609
Claims of Instant Application 19307558
Claims of Application 16707814
Claim 1. A computer-implemented method, comprising:
Claim 1. A computer-implemented method, comprising:
calling at least one artificial intelligence (AI) model, by an automation, when executing a probabilistic activity of a workflow;
calling at least one machine learning (ML) model, by a robotic process automation (RPA) robot, when executing a probabilistic activity of an RPA workflow;
receiving, by the automation, at least one confidence value from the at least one AI model;
receiving, by the RPA robot, at least one confidence value from the at least one ML model;
responsive to the at least one confidence value not exceeding a confidence threshold:
responsive to the at least one confidence value not exceeding a confidence threshold:
turning off a workflow section after the probabilistic activity, by the automation, via the probabilistic activity; and
turning off a workflow section after the probabilistic activity, by the RPA robot, via the probabilistic activity; and
responsive to the at least one confidence value exceeding the confidence threshold:
responsive to the at least one confidence value exceeding the confidence threshold:
turning on the workflow section after the probabilistic activity, by the automation, via the probabilistic activity, and
turning on the workflow section after the probabilistic activity, by the RPA robot, via the probabilistic activity, and
executing the workflow section following the probabilistic activity, by the automation, wherein the automation, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold, and
executing the workflow section following the probabilistic activity, by the RPA robot, wherein the RPA robot, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold, and
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the workflow.
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the RPA workflow.
Claim 2. The computer-implemented method of claim 1, further comprising:
Claim 2. The computer-implemented method of claim 1, further comprising:
generating the workflow comprising a plurality of deterministic activities and the at least one probabilistic activity configured to call the at least one AI model; and
generating the RPA workflow comprising a plurality of deterministic activities and the at least one probabilistic activity configured to call the at least one ML model; and
generating the automation that implements the generated workflow.
generating the RPA robot that implements the generated RPA workflow.
Claim 3. The computer-implemented method of claim 1, further comprising:
Claim 3. The computer-implemented method of claim 1, further comprising:
raising or lowering the confidence threshold, by the automation, after the process of the workflow has been run a predetermined number of times.
raising or lowering the confidence threshold, by the RPA robot, after the process of the workflow has been run a predetermined number of times.
Claim 4. The computer-implemented method of claim 3, wherein the raising or lowering of the confidence threshold by the automation is repeated until a winning state is achieved.
Claim 4. The computer-implemented method of claim 3, wherein the raising or lowering of the confidence threshold by the RPA robot is repeated until a winning state is achieved.
Claim 5. The computer-implemented method of claim 3, further comprising:
Claim 5. The computer-implemented method of claim 3, further comprising:
determining, by the automation, that the at least one AI model is not achieving an outcome after a predetermined number of modifications to the confidence threshold;
determining, by the RPA robot, that the at least one ML models are not achieving an outcome after a predetermined number of modifications to the confidence threshold; and
and retraining the at least one AI model.
retraining the at least one ML models.
Claim 6. The computer-implemented method of claim 1, wherein the automation is configured to determine how much of a certain resource to provide, determine a more optimal confidence threshold, or both.
Claim 6. The computer-implemented method of claim 1, wherein the RPA robot is configured to determine how much of a certain resource to provide, determine a more optimal confidence threshold, or both.
Claim 7. The computer-implemented method of claim 1, wherein the automation calls multiple AI models and the confidence values from each AI model are combined to determine a global confidence value that is compared against the confidence threshold for the probabilistic activity.
Claim 7. The computer-implemented method of claim 1, wherein the RPA robot calls multiple ML models and the confidence values from each ML model are combined to determine a global confidence value that is compared against the confidence threshold for the probabilistic activity.
Claim 8. The computer-implemented method of claim 7, wherein the global confidence value is determined by applying a respective weight to the confidence values and combining the weighted confidence values.
Claim 8. The computer-implemented method of claim 7, wherein the global confidence value is determined by applying a respective weight to the confidence values and combining the weighted confidence values.
Claim 9. One or more non-transitory computer-readable media storing one or more computer programs, the one or more computer programs configured to cause at least one processor to:
Claim 9. A non-transitory computer-readable medium storing a computer program, the computer program configured to cause at least one processor to:
call an artificial intelligence (AI) model while executing a probabilistic activity of a workflow;
call a machine learning (ML) model while executing a probabilistic activity of an RPA workflow;
receive a confidence value from the AI model;
receive a confidence value from the ML model;
responsive to the confidence value not exceeding a confidence threshold, turn off a workflow section after the probabilistic activity, via the probabilistic activity; and
responsive to the confidence value not exceeding a confidence threshold, turn off a workflow section after the probabilistic activity, via the probabilistic activity; and
responsive to the confidence value exceeding the confidence threshold:
responsive to the confidence value exceeding the confidence threshold:
turn on a workflow section after the probabilistic activity, via the probabilistic activity, and
turn on a workflow section after the probabilistic activity, via the probabilistic activity, and
execute the workflow section following the probabilistic activity, wherein an automation, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold, and
execute the workflow section following the probabilistic activity, wherein the RPA robot, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold, and
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the workflow.
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the RPA workflow.
Claim 10. The one or more non-transitory computer-readable media of claim 9, wherein the one or more computer programs are further configured to cause the at least one processor to:
Claim 10. The non-transitory computer-readable medium of claim 9, wherein the computer program is further configured to cause the at least one processor to:
raise or lower the confidence threshold after the process of the workflow has been run a predetermined number of times.
raise or lower the confidence threshold after the process of the workflow has been run a predetermined number of times.
Claim 11. The one or more non-transitory computer-readable media of claim 10, wherein the raising or lowering of the confidence threshold is repeated until a winning state is achieved.
Claim 11. The non-transitory computer-readable medium of claim 10, wherein the raising or lowering of the confidence threshold is repeated until a winning state is achieved.
Claim 12. The one or more non-transitory computer-readable media of claim 10, wherein the one or more computer programs are further configured to cause the at least one processor to:
Claim 12. The non-transitory computer-readable medium of claim 10, wherein the computer program is further configured to cause the at least one processor to:
determine that the AI model is not achieving an outcome after a predetermined number of modifications to the confidence threshold; and
determine that the ML model is not achieving an outcome after a predetermined number of modifications to the confidence threshold; and
provide an indication to a server to retrain the AI model.
provide an indication to a server to retrain the ML model.
Claim 13. The one or more non-transitory computer-readable media of claim 9, wherein the one or more computer programs are further configured to cause the at least one processor to:
Claim 13. The non-transitory computer-readable medium of claim 9, wherein the computer program is further configured to cause the at least one processor to:
determine how much of a certain resource to provide, determine a more optimal confidence threshold, or both.
determine how much of a certain resource to provide, determine a more optimal confidence threshold, or both.
Claim 14. A computing system, comprising:
Claim 14. A computer-implemented method, comprising:
memory storing computer program instructions; and
at least one processor configured to execute the computer program instructions, wherein the computer program instructions are configured to cause the at least one processor to:
call at least one artificial intelligence (AI) model while executing a probabilistic activity of a workflow,
calling at least one machine learning (ML) model, by a robotic process automation (RPA) robot, while executing a probabilistic activity of an RPA workflow;
receive at least one confidence value from the at least one AI model,
receiving, by the RPA robot, at least one confidence value from the at least one ML model;
compare the at least one confidence value to a plurality of confidence threshold ranges,
comparing the at least one confidence value to a plurality of confidence threshold ranges, by the RPA robot;
modify the confidence threshold ranges based on an applied scenario, and
modifying the confidence threshold ranges based on an applied scenario; and
responsive to the at least one confidence value falling within a confidence threshold range:
responsive to the at least one confidence value falling within a confidence threshold range:
turn on a workflow section after the probabilistic activity for that confidence threshold range via the probabilistic activity, and
turning on a workflow section after the probabilistic activity for that confidence threshold range, by the RPA robot, via the probabilistic activity, and
execute the workflow section following the probabilistic activity for that confidence threshold range, wherein an automation, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold, and
executing the workflow section following the probabilistic activity for that confidence threshold range, by the RPA robot, wherein the RPA robot, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold, and
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the workflow.
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the RPA workflow.
Claim 15. The one or more computing systems of claim 14, wherein the computer program instructions are further configured to cause the at least one processor to:
Claim 15. The computer-implemented method of claim 14, further comprising:
generate the workflow comprising a plurality of deterministic activities and the at least one probabilistic activity configured to call the at least one AI model; and
generating the RPA workflow comprising a plurality of deterministic activities and the at least one probabilistic activity configured to call the at least one ML model; and
generate the automation that implements the generated workflow.
generating the RPA robot that implements the generated RPA workflow.
Claim 16. The one or more computing systems of claim 14, wherein the computer program instructions are further configured to cause the at least one processor to:
Claim 16. The computer-implemented method of claim 14, further comprising:
modify one or more of the confidence threshold ranges after the process of the workflow is run a predetermined number of times.
modifying one or more of the confidence threshold ranges, by the RPA robot, after the process of the workflow is run a predetermined number of times.
Claim 17. The one or more computing systems of claim 16, wherein the modification of the one or more confidence threshold ranges is repeated until a winning state is achieved.
Claim 17. The computer-implemented method of claim 16, wherein the modification of the one or more confidence threshold ranges by the RPA robot is repeated until a winning state is achieved.
Claim 18. The one or more computing systems of claim 16, wherein the computer program instructions are further configured to cause the at least one processor to:
Claim 18. The computer-implemented method of claim 16, further comprising:
determine that the at least one AI model is not achieving an outcome after a predetermined number of modifications to the confidence threshold range; and
determining, by the RPA robot, that the at least one ML models are not achieving an outcome after a predetermined number of modifications to the confidence threshold range; and
retrain the at least one AI model.
retraining the at least one ML models.
Claim 19. The one or more computing systems of claim 14, wherein the automation is configured to determine how much of a certain resource to provide, determine a more optimal confidence threshold range, or both.
Claim 19. The computer-implemented method of claim 14, wherein the RPA robot is configured to determine how much of a certain resource to provide, determine a more optimal confidence threshold range, or both.
Claim 20. The one or more computing systems of claim 14, wherein the RPA robot calls multiple AI models and the confidence values from each AI model are combined to determine a global confidence value that is compared against the confidence threshold ranges for the probabilistic activity, and
Claim 20. The computer-implemented method of claim 14, wherein the RPA robot calls multiple ML models and the confidence values from each ML model are combined to determine a global confidence value that is compared against the confidence threshold ranges for the probabilistic activity, and
the global confidence value is determined by applying a respective weight to the confidence values and combining the weighted confidence values.
the global confidence value is determined by applying a respective weight to the confidence values and combining the weighted confidence values.
Claim Rejections - 35 USC § 102
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)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1, 2, 6, 9, and 13 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by S NANAL et al. (US 20200180148 A1, hereinafter Nanal).
Regarding claim 1, Nanal discloses the computer-implemented method, comprising:
calling at least one artificial intelligence (AI) model, by an automation, when executing a probabilistic activity of a workflow (at least as in paragraph 0045, wherein “At 410, the scores of the existing solutions may be compared to determine if one or more of the existing solutions indicate a match with the issue”; at least as in paragraph 0033, wherein “Custom machine learning (ML) based methodologies are used to identify the correct solutions for a given issue and update the knowledge base 108 in accordance with the examples disclosed herein”; at least as in paragraph 0021, wherein “The RPA system detects fallouts/issues and resolves the issues using cognitive intelligence developed based on machine learning (ML) and natural language processing techniques (NLP)”);
receiving, by the automation, at least one confidence value from the at least one AI model (at least as in paragraph 0029 and 0038, wherein “data analyzer 106 includes a solution evaluator 302, an issue categorizer 304, a bot selector 306 and a bot generator 308 . . . the solution evaluator 302 evaluates the various solutions 312 . . . the solution evaluator 302 can implement a decision tree methodology for generating scores 322 for the solutions 312 based on matches between the features of the issue 152 and the features of each of the solutions”; at least as in paragraph 0050, wherein “The decision tree methodology enables identifying the correct solution for the issue arising . . . via traversing the tree paths or branches of the tree structure 700 . . . Each feature of the solution may be associated with a respective score based on a match between the solution feature and the issue features”);
responsive to the at least one confidence value not exceeding a confidence threshold:
turning off a workflow section after the probabilistic activity, by the automation, via the probabilistic activity (at least as in paragraph 0045, wherein “The existing solutions are evaluated at 408 for matching with the issue 152 . . . a nearest neighbor methodology can be adapted for evaluating the existing solutions wherein various features of the issue 152 are checked for match against the existing solutions . . . If it is determined at 412 that no solutions exist for implementing to resolve the issue 152, then the issue is presented to the user for review at 414”); and
responsive to the at least one confidence value exceeding the confidence threshold:
turning on the workflow section after the probabilistic activity, by the automation, via the probabilistic activity, and
executing the workflow section following the probabilistic activity, by the automation (at least as in paragraph 0045, wherein “If it is determined at 410, that the one or more solutions indicate a match, then such solutions are selected at 412 for implementation to resolve the issue 152”; at least as in paragraph 0039, wherein “scores 322 are used by the issue categorizer 304 to categorize the issue 152 as one of a new issue or an existing issue . . . the issue categorizer 304 can include classifiers that are trained to select particular solutions for the issue 152 based on the scores 322”; at least as in paragraph 0050, wherein “Based on the respective overall scores of various branches for the various features associated with the leaves of the branch, a particular branch of the tree structure 700 having a highest score may be selected as the correct solution. The overall respective scores of the tree paths can be compared with a predetermined score threshold for the selection of the particular branch”),
wherein the automation, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold (at least as in paragraph 0042, wherein “When the issue 152 is categorized as a new issue upon the implementation of the decision tree methodology for the score evaluation, the bot selector 306 may execute a further check on the solutions 312 using a matching methodology such as the nearest neighbor . . . The confidence levels associated with the prior solutions 312 may be evaluated to determine if any solutions can be selected to resolve the issue 152”), and
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the workflow (at least as in paragraph 0045 and 0046, wherein “The user actions to resolve the issue are recorded at 416 and a solution associated with the recorded user actions is implemented within the application stack 150 at 420 to resolve the issue 152. The recorded user actions and the results of the recorded user actions which can include the success or failure of the recorded user actions in resolving the issue may be updated to the knowledge base 108 at 422. If the recorded user actions result in resolving the issue 152 thereby ensuring a smooth, error-free processing of the entity 110 through the application stack 150, the user actions may be added to the existing solutions and made available for implementation when an issue arises during the processing of an entity through the application stack 150”).
Regarding claim 2, Nanal further discloses the computer-implemented method of claim 1, further comprising:
generating the workflow comprising a plurality of deterministic activities and the at least one probabilistic activity configured to call the at least one AI model (at least as in paragraph 0043-0045 and Fig. 4, wherein the method of implementing a robotic process automation includes receiving information regarding the issue leading to reformulating the error messages, and further wherein the method includes determining whether the issue is a new issue and evaluating and using existing solutions); and
generating the automation that implements the generated workflow (at least as in paragraph 0053-0055, wherein the RPA system 100 is implemented on a computer system; at least as in paragraph 0043, wherein the RPA system implements the method of robotic process automation; therefore, the RPA system implementing the method – or workflow – of RPA is implemented).
Regarding claim 6, Nanal further discloses the computer-implemented method of claim 1, wherein the automation is configured to determine how much of a certain resource to provide, determine a more optimal confidence threshold, or both (at least as in paragraph 0050, wherein “The decision tree methodology enables identifying the correct solution for the issue arising . . . via traversing the tree paths or branches of the tree structure 700 . . . Each feature of the solution may be associated with a respective score based on a match between the solution feature and the issue features”; at least as in paragraph 0045, wherein “If it is determined at 410, that the one or more solutions indicate a match, then such solutions are selected at 412 for implementation to resolve the issue 152”; at least as in paragraph 0039, wherein “scores 322 are used by the issue categorizer 304 to categorize the issue 152 as one of a new issue or an existing issue . . . the issue categorizer 304 can include classifiers that are trained to select particular solutions for the issue 152 based on the scores 322”; therefore, the RPA system determines which solution to apply).
Regarding claim 9, Nanal discloses:
One or more non-transitory computer-readable media storing one or more computer programs, the one or more computer programs (at least as in paragraph 0055, wherein “RPA system 100 may be implemented as software stored on a non-transitory computer-readable medium and executed by the one or more processors 902”), the one or more computer programs configured to cause at least one processor to:
call an artificial intelligence (AI) model while executing a probabilistic activity of a workflow (at least as in paragraph 0045, wherein “At 410, the scores of the existing solutions may be compared to determine if one or more of the existing solutions indicate a match with the issue”; at least as in paragraph 0033, wherein “Custom machine learning (ML) based methodologies are used to identify the correct solutions for a given issue and update the knowledge base 108 in accordance with the examples disclosed herein”; at least as in paragraph 0021, wherein “The RPA system detects fallouts/issues and resolves the issues using cognitive intelligence developed based on machine learning (ML) and natural language processing techniques (NLP)”);
receive a confidence value from the AI model (at least as in paragraph 0029 and 0038, wherein “data analyzer 106 includes a solution evaluator 302, an issue categorizer 304, a bot selector 306 and a bot generator 308 . . . the solution evaluator 302 evaluates the various solutions 312 . . . the solution evaluator 302 can implement a decision tree methodology for generating scores 322 for the solutions 312 based on matches between the features of the issue 152 and the features of each of the solutions”; at least as in paragraph 0050, wherein “The decision tree methodology enables identifying the correct solution for the issue arising . . . via traversing the tree paths or branches of the tree structure 700 . . . Each feature of the solution may be associated with a respective score based on a match between the solution feature and the issue features”);
responsive to the confidence value not exceeding a confidence threshold, turn off a workflow section after the probabilistic activity, via the probabilistic activity (at least as in paragraph 0045, wherein “The existing solutions are evaluated at 408 for matching with the issue 152 . . . a nearest neighbor methodology can be adapted for evaluating the existing solutions wherein various features of the issue 152 are checked for match against the existing solutions . . . If it is determined at 412 that no solutions exist for implementing to resolve the issue 152, then the issue is presented to the user for review at 414”); and
responsive to the confidence value exceeding the confidence threshold:
turn on a workflow section after the probabilistic activity, via the probabilistic activity, and
execute the workflow section following the probabilistic activity (at least as in paragraph 0045, wherein “If it is determined at 410, that the one or more solutions indicate a match, then such solutions are selected at 412 for implementation to resolve the issue 152”; at least as in paragraph 0039, wherein “scores 322 are used by the issue categorizer 304 to categorize the issue 152 as one of a new issue or an existing issue . . . the issue categorizer 304 can include classifiers that are trained to select particular solutions for the issue 152 based on the scores 322”; at least as in paragraph 0050, wherein “Based on the respective overall scores of various branches for the various features associated with the leaves of the branch, a particular branch of the tree structure 700 having a highest score may be selected as the correct solution. The overall respective scores of the tree paths can be compared with a predetermined score threshold for the selection of the particular branch”), wherein
an automation, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold (at least as in paragraph 0042, wherein “When the issue 152 is categorized as a new issue upon the implementation of the decision tree methodology for the score evaluation, the bot selector 306 may execute a further check on the solutions 312 using a matching methodology such as the nearest neighbor . . . The confidence levels associated with the prior solutions 312 may be evaluated to determine if any solutions can be selected to resolve the issue 152”), and
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the workflow (at least as in paragraph 0045 and 0046, wherein “The user actions to resolve the issue are recorded at 416 and a solution associated with the recorded user actions is implemented within the application stack 150 at 420 to resolve the issue 152. The recorded user actions and the results of the recorded user actions which can include the success or failure of the recorded user actions in resolving the issue may be updated to the knowledge base 108 at 422. If the recorded user actions result in resolving the issue 152 thereby ensuring a smooth, error-free processing of the entity 110 through the application stack 150, the user actions may be added to the existing solutions and made available for implementation when an issue arises during the processing of an entity through the application stack 150”).
Regarding claim 13, Nanal further discloses the one or more one or more non-transitory computer-readable media of claim 9, wherein the one or more computer programs are further configured to cause the at least one processor to:
determine how much of a certain resource to provide, determine a more optimal confidence threshold, or both (at least as in paragraph 0050, wherein “The decision tree methodology enables identifying the correct solution for the issue arising . . . via traversing the tree paths or branches of the tree structure 700 . . . Each feature of the solution may be associated with a respective score based on a match between the solution feature and the issue features”; at least as in paragraph 0045, wherein “If it is determined at 410, that the one or more solutions indicate a match, then such solutions are selected at 412 for implementation to resolve the issue 152”; at least as in paragraph 0039, wherein “scores 322 are used by the issue categorizer 304 to categorize the issue 152 as one of a new issue or an existing issue . . . the issue categorizer 304 can include classifiers that are trained to select particular solutions for the issue 152 based on the scores 322”; therefore, the RPA system determines which solution to apply).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 3, 10, 14, 15, 16, and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over S Nanal et al. (US 20200180148 A1, hereinafter Nanal) in view of Wu et al (US 2019/0205761 A1, hereinafter Wu), and further in view of Maruyama (US 20200333777 A1).
Regarding claim 3, Nanal further discloses the computer-implemented method of claim 1, further comprising:
[comparing or matching the predetermined score threshold], by the automation (at least as in paragraph 0050, wherein the RPA system compares the scores of tree paths with the predetermined score threshold). But Nanal does not specifically teach “raising or lowering the confidence threshold … after the process of the workflow has been run a predetermined number of times.”
However, Wu discloses a neural-network based mechanism for providing a human intermediary for intervening in searches when there is low confidence that pre-existing categories map to a user’s search string. Wu specifically discloses wherein the “confidence threshold may be dynamically determined based on a number of human resources available or expected to be available at a particular point in time” (at least as in paragraphs 0012, 0075, 0223). “For example, if there are a larger amount of resources available (e.g., ten agents), the model rejector component of neural network 212 may apply a higher threshold for confidence for automatic classification, and if there are less resources available (e.g., one agent), the model rejector component of neural network 212 may apply a lower threshold for confidence for automatic classification” (at least as in paragraph 0075). Thus, Wu teaches “raising or lowering the confidence threshold.”
Maruyama discloses an abnormality detection apparatus acquires observation values serving as indexes of an operating state of a monitoring target apparatus at predetermined timings in a process executed repeatedly in the monitoring target apparatus. Maruyama specifically discloses wherein one run of a process may be a process for a batch in a batch process or a process for a wafer in a sheet process (at least as in paragraph 0068). Maruyama further discloses wherein the semiconductor manufacturing apparatus transmits the observation value, or a value indicating the operating state of the step at predetermined timing of the step executed in the semiconductor manufacturing apparatus, to the observation value acquisition unit when the same process is repeated a predetermined number of times (at least as in paragraphs 0067 and 0068). Thus, Maruyama teaches “after the process of the workflow has been run a predetermined number of times.”
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include Wu’s teachings of a neural-network based system dynamically determining and modifying thresholds and Maruyama’s teaching of transmitting the observation value when the same process is repeated a predetermined number of times, since Wu teaches wherein such system improve the performance of the neural network in terms of computational speed and resource usage and Maruyama teaches wherein transmission of a set of observation values allows for bulk processing, thus minimizing the number of data transmissions required in between units, and generation of summary values for the set of observation values.
Regarding claim 10, Nanal further discloses the one or more non-transitory computer-readable media of claim 9, wherein the one or more computer programs are further configured to cause the at least one processor to:
[comparing or matching the predetermined score threshold], by the automation (at least as in paragraph 0050, wherein the RPA system compares the scores of tree paths with the predetermined score threshold). But Nanal does not specifically teach “raise or lower the confidence threshold … after the process of the workflow has been run a predetermined number of times.”
However, Wu discloses a neural-network based mechanism for providing a human intermediary for intervening in searches when there is low confidence that pre-existing categories map to a user’s search string. Wu specifically discloses wherein the “confidence threshold may be dynamically determined based on a number of human resources available or expected to be available at a particular point in time” (at least as in paragraphs 0012, 0075, 0223). “For example, if there are a larger amount of resources available (e.g., ten agents), the model rejector component of neural network 212 may apply a higher threshold for confidence for automatic classification, and if there are less resources available (e.g., one agent), the model rejector component of neural network 212 may apply a lower threshold for confidence for automatic classification” (at least as in paragraph 0075). Thus, Wu teaches “raise or lower the confidence threshold.”
Maruyama discloses an abnormality detection apparatus acquires observation values serving as indexes of an operating state of a monitoring target apparatus at predetermined timings in a process executed repeatedly in the monitoring target apparatus. Maruyama specifically discloses wherein one run of a process may be a process for a batch in a batch process or a process for a wafer in a sheet process (at least as in paragraph 0068). Maruyama further discloses wherein the semiconductor manufacturing apparatus transmits the observation value, or a value indicating the operating state of the step at predetermined timing of the step executed in the semiconductor manufacturing apparatus, to the observation value acquisition unit when the same process is repeated a predetermined number of times (at least as in paragraphs 0067 and 0068). Thus, Maruyama teaches “after the process of the workflow has been run a predetermined number of times.”
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include Wu’s teachings of a neural-network based system dynamically determining and modifying thresholds and Maruyama’s teaching of transmitting the observation value when the same process is repeated a predetermined number of times, since Wu teaches wherein such system improve the performance of the neural network in terms of computational speed and resource usage and Maruyama teaches wherein transmission of a set of observation values allows for bulk processing, thus minimizing the number of data transmissions required in between units, and generation of summary values for the set of observation values.
Regarding claim 14, Nanal discloses:
A computing system (Fig. 9, computer system 900), comprising:
memory storing computer program instructions (at least as in paragraph 0054 & Fig. 9, “the computer-readable medium 906 may be non-transitory or non-volatile medium, such as a magnetic disk or solid-state non-volatile memory or volatile medium such as RAM. The instructions or modules stored on the computer-readable medium 906 may include machine-readable instructions 964 executed by the processor(s) 902 to perform the methods and functions of the RPA system 100”); and
at least one processor configured to execute the computer program instructions (at least as in paragraph 0054, “The computer system 900 includes processor(s) 902, such as a central processing unit, ASIC or other type of processing circuit”), wherein the computer program instructions are configured to cause the at least one processor to:
call at least one artificial intelligence (AI) model, by an automation, while executing a probabilistic activity of a workflow (at least as in paragraph 0045, wherein “At 410, the scores of the existing solutions may be compared to determine if one or more of the existing solutions indicate a match with the issue”; at least as in paragraph 0033, wherein “Custom machine learning (ML) based methodologies are used to identify the correct solutions for a given issue and update the knowledge base 108 in accordance with the examples disclosed herein”; at least as in paragraph 0021, wherein “The RPA system detects fallouts/issues and resolves the issues using cognitive intelligence developed based on machine learning (ML) and natural language processing techniques (NLP)”);
receive at least one confidence value from the at least one AI model (at least as in paragraph 0029 and 0038, wherein “data analyzer 106 includes a solution evaluator 302, an issue categorizer 304, a bot selector 306 and a bot generator 308 . . . the solution evaluator 302 evaluates the various solutions 312 . . . the solution evaluator 302 can implement a decision tree methodology for generating scores 322 for the solutions 312 based on matches between the features of the issue 152 and the features of each of the solutions”; at least as in paragraph 0050, wherein “The decision tree methodology enables identifying the correct solution for the issue arising . . . via traversing the tree paths or branches of the tree structure 700 . . . Each feature of the solution may be associated with a respective score based on a match between the solution feature and the issue features”; at least as in paragraph 0039, wherein “The scores 322 are used by the issue categorizer 304 to categorize the issue 152 as one of a new issue or an existing issue;”);
compare the at least one confidence value at least as in paragraph 0042, wherein “The confidence levels associated with the prior solutions 312 may be evaluated to determine if any solutions can be selected to resolve the issue 152;” at least as in paragraph 0045, wherein “the scores of the existing solutions may be compared to determine if one or more of the existing solutions indicate a match with the issue”);
responsive to the at least one confidence value falling within a confidence threshold range:
turn on a workflow section after the probabilistic activity for that confidence threshold range via the probabilistic activity, and
execute the workflow section following the probabilistic activity for that confidence threshold range, (at least as in paragraph 0042, wherein “If none of the prior solutions 312 are selected, then the issue 152 is flagged for user review;” at least as in paragraph 0045, wherein “If it is determined at 410, that the one or more solutions indicate a match, then such solutions are selected at 412 for implementation to resolve the issue 152”; at least as in paragraph 0039, wherein “scores 322 are used by the issue categorizer 304 to categorize the issue 152 as one of a new issue or an existing issue . . . the issue categorizer 304 can include classifiers that are trained to select particular solutions for the issue 152 based on the scores 322”; at least as in paragraph 0050, wherein “Based on the respective overall scores of various branches for the various features associated with the leaves of the branch, a particular branch of the tree structure 700 having a highest score may be selected as the correct solution. The overall respective scores of the tree paths can be compared with a predetermined score threshold for the selection of the particular branch”), wherein an automation, in the probabilistic activity, determines whether the at least one confidence value exceeds the confidence threshold, and (at least as in paragraph 0042, wherein “When the issue 152 is categorized as a new issue upon the implementation of the decision tree methodology for the score evaluation, the bot selector 306 may execute a further check on the solutions 312 using a matching methodology such as the nearest neighbor . . . The confidence levels associated with the prior solutions 312 may be evaluated to determine if any solutions can be selected to resolve the issue 152”; at least as in paragraph 0050, wherein “Each feature of the solution may be associated with a respective score based on a match between the solution feature and the issue features. Based on the respective overall scores of various branches for the various features associated with the leaves of the branch, a particular branch of the tree structure 700 having a highest score may be selected as the correct solution. The overall respective scores of the tree paths can be compared with a predetermined score threshold for the selection of the particular branch”), and
the turning on or off of the workflow section after the probabilistic activity comprises modifying activity parameters within the workflow (at least as in paragraph 0045 and 0046, wherein “The user actions to resolve the issue are recorded at 416 and a solution associated with the recorded user actions is implemented within the application stack 150 at 420 to resolve the issue 152. The recorded user actions and the results of the recorded user actions which can include the success or failure of the recorded user actions in resolving the issue may be updated to the knowledge base 108 at 422. If the recorded user actions result in resolving the issue 152 thereby ensuring a smooth, error-free processing of the entity 110 through the application stack 150, the user actions may be added to the existing solutions and made available for implementation when an issue arises during the processing of an entity through the application stack 150”; at least as in paragraph 0040, wherein “When the issue 152 is categorized as an existing issue, the data regarding the issue 152 e.g., the application at which the data fallout occurred, the type of error that occurred and the solution 354 which was selected from the prior solutions 312 for resolving the issue 152 is transmitted to the bot selector 306. The bot selector 306 can identify and select one or more of the bots 126 for implementing the selected solution 354 at one or more of the applications application_1, application_2 . . . application_n in the application stack 150”).
However, Nanal does not explicitly disclose “to a plurality of confidence threshold ranges” and “modify the confidence threshold ranges based on an applied scenario.”
Wu discloses a neural-network based mechanism for providing a human intermediary for intervening in searches when there is low confidence that pre-existing categories map to a user’s search string. Wu specifically discloses wherein the “confidence threshold may be dynamically determined based on a number of human resources available or expected to be available at a particular point in time” (at least as in paragraphs 0012, 0075, 0223). “For example, if there are a larger amount of resources available (e.g., ten agents), the model rejector component of neural network 212 may apply a higher threshold for confidence for automatic classification, and if there are less resources available (e.g., one agent), the model rejector component of neural network 212 may apply a lower threshold for confidence for automatic classification” (at least as in paragraph 0075). Wu further discloses wherein “weights are still updated in additional training in later stages” (at least as in paragraph 0144), and further wherein the model training process “picks a batch of training example pairs, runs a forward pass, and gets the signal to run the backpropagation” (at least as in paragraph 0216). Wu additionally discloses wherein “the processor is configured to re-train the neural network with the selected candidate category of the one or more candidate categories as a labelled training data element, adjusting weights within connected nodes of the neural network to minimize a loss function” (at least as in claim 4). Thus, Wu teaches “modify the confidence threshold ranges based on an applied scenario, and.”
Maruyama discloses an abnormality detection apparatus that acquires and determines a confidence or predictive value through the use of machine learning and other predictive models to detect the presence or absence of abnormality of the monitoring target apparatus. Maruyama specifically discloses wherein the abnormality detection method utilizes the upper and lower limit thresholds “UCL2” and “LCL2” in addition to the upper and lower limit thresholds “UCL1” and “LCL1” (at least as in paragraph 0126-0127). Thus, Maruyama teaches “to a plurality of confidence threshold ranges.”
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include Wu’s teachings of a neural-network based system dynamically determining and modifying parameters and retraining and Maruyama’s teaching of threshold ranges, since Wu teaches wherein such system improve the performance of the neural network in terms of computational speed and resource usage and Maruyama teaches wherein the threshold ranges have an effect of achieving accurate and efficient abnormality detection.
Regarding claim 15, in view of the above combination of Nanal, Wu, and Maruyama, Nanal further discloses the one or more computing systems of claim 14, further comprising:
generating the workflow comprising a plurality of deterministic activities and the at least one probabilistic activity configured to call the at least one AI model (at least as in paragraph 0043-0045 and Fig. 4, wherein the method of implementing a robotic process automation includes receiving information regarding the issue leading to reformulating the error messages, and further wherein the method includes determining whether the issue is a new issue and evaluating and using existing solutions); and
generating the automation that implements the generated workflow (at least as in paragraph 0053-0055, wherein the RPA system 100 is implemented on a computer system; at least as in paragraph 0043, wherein the RPA system implements the method of robotic process automation; therefore, the RPA system implementing the method – or workflow – of RPA is implemented).
Regarding claim 16, in view of the above combination of Nanal, Wu, and Maruyama, Nanal further discloses the one or more computing systems of claim 14, further comprising:
at least as in paragraph 0056, wherein “he data storage 910 may be used to store the entity information, entity features that were extracted, the solutions and the scores for the solutions during the issue-score matching process and the like”).
However, Nanal does not explicitly disclose wherein “modifying one or more of the confidence threshold ranges” and “a predetermined number of times.”
Wu discloses a neural-network based mechanism for providing a human intermediary for intervening in searches when there is low confidence that pre-existing categories map to a user’s search string. Wu specifically discloses wherein the “confidence threshold may be dynamically determined based on a number of human resources available or expected to be available at a particular point in time” (at least as in paragraphs 0012, 0075, 0223). “For example, if there are a larger amount of resources available (e.g., ten agents), the model rejector component of neural network 212 may apply a higher threshold for confidence for automatic classification, and if there are less resources available (e.g., one agent), the model rejector component of neural network 212 may apply a lower threshold for confidence for automatic classification” (at least as in paragraph 0075). Thus, Wu teaches “modifying a confidence threshold range.”
Maruyama discloses an abnormality detection apparatus that acquires and determines a confidence or predictive value through the use of machine learning and other predictive models to detect the presence or absence of abnormality of the monitoring target apparatus. Maruyama specifically discloses wherein the abnormality detection method utilizes the upper and lower limit thresholds “UCL2” and “LCL2” in addition to the upper and lower limit thresholds “UCL1” and “LCL1” (at least as in paragraph 0126-0127), and further wherein “the abnormality detection apparatus 1 dynamically changes the upper and the lower limit thresholds” (at least as in paragraph 0129). Maruyama further discloses wherein one run of a process may be a process for a batch in a batch process or a process for a wafer in a sheet process (at least as in paragraph 0068). Maruyama additionally discloses wherein the semiconductor manufacturing apparatus transmits the observation value, or a value indicating the operating state of the step at predetermined timing of the step executed in the semiconductor manufacturing apparatus, to the observation value acquisition unit when the same process is repeated a predetermined number of times (at least as in paragraphs 0067 and 0068).
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include Wu’s teachings of a neural-network based system dynamically determining and modifying parameters and Maruyama’s teaching of dynamically changing numerous upper and lower limit thresholds and transmitting the observation value when the same process is repeated a predetermined number of times, since Wu teaches wherein such system improve the performance of the neural network in terms of computational speed and resource usage and Maruyama teaches wherein the threshold ranges, dynamic adaptation, and detection transmission have an effect of achieving accurate and efficient abnormality detection.
Regarding claim 19, the above combination of Nanal, Wu, and Maruyama teaches the one or more computing systems of claim 14, however, Nanal does not specifically teach “wherein the automation is configured to determine how much of a certain resource to provide, determine a more optimal confidence threshold, or both.”
Wu discloses a neural-network based mechanism for providing a human intermediary for intervening in searches when there is low confidence that pre-existing categories map to a user’s search string. Wu additionally teaches wherein said mechanism can modify the decision to submit a user’s query to said human intermediary by dynamically determining the confidence threshold based on a number of human resources available or expected to be available at a particular point in time (at least as in paragraphs 0012 & 0075).
Therefore, it would have been obvious to one of ordinary skill in the art at the effective filing date of the claimed invention to modify the teachings of Nanal, to include Wu’s teachings of a neural-network based system dynamically determining and modifying parameters such as the confidence threshold required based on available resources, since Wu teaches wherein such system improve the performance of the neural network in terms of computational speed and resource usage.
Claim(s) 4, 5, 11, 12, and 17-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over S Nanal et al. (US 20200180148 A1, hereinafter Nanal) in view of Wu et al (US 2019/0205761 A1, hereinafter Wu) and Maruyama (US 20200333777 A1), and further in view of He et al (US 2017/0082555 A1, hereinafter He).
Regarding claim 4, in view of the above combination of Nanal, Wu, and Maruyama, Nanal further discloses: The computer-implemented method of claim 3, wherein [comparing or matching the predetermined score threshold] by the automation (at least as in paragraph 0050, wherein the RPA system compares the scores of tree paths with the predetermined score threshold, selects a correct solution, etc.).
However, Nanal does not explicitly disclose wherein “raising or lowering of the confidence threshold” and “until a winning state is achieved.”
Wu discloses a neural-network based mechanism for providing a human intermediary for intervening in searches when there is low confidence that pre-existing categories map to a user’s search string. Wu specifically discloses wherein the “confidence threshold may be dynamically determined based on a number of human resources available or expected to be available at a particular point in time” (at least as in paragraphs 0012, 0075, 0223). “For example, if there are a larger amount of resources available (e.g., ten agents), the model rejector component of neural network 212 may apply a higher threshold for confidence for automatic classification, and if there are less resources available (e.g., one agent), the model rejector component of neural network 212 may apply a lower threshold for confidence for automatic classification” (at least as in paragraph 0075). Thus, Wu teaches “raising or lowering of the confidence threshold.”
He discloses a system for classifying defects detected during the semiconductor fabrication process wherein the defect classifier utilizing machine learning is modified using a training set that includes the defects classified with a confidence below a threshold. He specifically discloses wherein an automated classifier may be created with auto confidence threshold that increases the confidence threshold from a minimum value until it reaches a target level of purity (at least as in paragraph 0053).
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include Wu’s teachings of a neural-network based system dynamically determining and modifying parameters and He’s teaching of reaching a purity target with an automatic confidence threshold, since Wu teaches wherein such system improve the performance of the neural network in terms of computational speed and resource usage and He teaches wherein the automatic confidence threshold improves the cost of ownership as the improved accuracy and reduced false positives require less manual review.
Regarding claim 5, in view of the above combination of Nanal, Wu, and Maruyama, Nanal further discloses: The computer-implemented method of claim 3, further comprising:
determining, by the automation, that the at least one AI models are not achieving an outcome … (at least as in paragraph 0045, wherein “If it is determined at 412 that no solutions exist for implementing to resolve the issue 152, then the issue is presented to the user for review at 414”); and
retraining the at least one AI models (at least as in paragraph 0042, wherein after the system determines no solution exists, the user may provide feedback with the correct solution for the new issue and “the user gestures 356 to fix the issue 152 can be recorded as a new solution within the knowledge base 108 and the bots for implementing the correct solution may be trained on the user gestures 356 by the bot generator 308”).
But Nanal does not specifically disclose “after a predetermined number of modifications to the confidence threshold.”
He discloses a system for classifying defects detected during the semiconductor fabrication process wherein the defect classifier utilizing machine learning is modified using a training set that includes the defects classified with a confidence below a threshold. He specifically discloses wherein an automated classifier may be created with auto confidence threshold that increases the confidence threshold from a minimum value until it reaches a target level of purity (at least as in paragraph 0053). He additionally discloses wherein the number of novel defects detected by the classifier exceeds a threshold (at least as in paragraphs 0092-0093 ). He further discloses wherein once the number of novel defect exceeds a threshold, re-training of the classifier may be triggered (at least as in paragraph 0092-0093).
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include He’s teaching of reaching a purity target with an automatic confidence threshold and retraining after a certain number of novel defects, since He teaches wherein the automatic confidence threshold and retraining improves accuracy and reduces false positives thus improving the cost of ownership and amount of manual review.
Regarding claim 11, in view of the above combination of Nanal, Wu, and Maruyama, Nanal further discloses: The one or more non-transitory computer-readable media of claim 10, wherein [comparing or matching the predetermined score threshold] by the automation (at least as in paragraph 0050, wherein the RPA system compares the scores of tree paths with the predetermined score threshold, selects a correct solution, etc.).
However, Nanal does not explicitly disclose wherein “raising or lowering of the confidence threshold” and “until a winning state is achieved.”
Wu discloses a neural-network based mechanism for providing a human intermediary for intervening in searches when there is low confidence that pre-existing categories map to a user’s search string. Wu specifically discloses wherein the “confidence threshold may be dynamically determined based on a number of human resources available or expected to be available at a particular point in time” (at least as in paragraphs 0012, 0075, 0223). “For example, if there are a larger amount of resources available (e.g., ten agents), the model rejector component of neural network 212 may apply a higher threshold for confidence for automatic classification, and if there are less resources available (e.g., one agent), the model rejector component of neural network 212 may apply a lower threshold for confidence for automatic classification” (at least as in paragraph 0075). Thus, Wu teaches “raising or lowering of the confidence threshold.”
He discloses a system for classifying defects detected during the semiconductor fabrication process wherein the defect classifier utilizing machine learning is modified using a training set that includes the defects classified with a confidence below a threshold. He specifically discloses wherein an automated classifier may be created with auto confidence threshold that increases the confidence threshold from a minimum value until it reaches a target level of purity (at least as in paragraph 0053).
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include Wu’s teachings of a neural-network based system dynamically determining and modifying parameters and He’s teaching of reaching a purity target with an automatic confidence threshold, since Wu teaches wherein such system improve the performance of the neural network in terms of computational speed and resource usage and He teaches wherein the automatic confidence threshold improves the cost of ownership as the improved accuracy and reduced false positives require less manual review.
Regarding claim 12, in view of the above combination of Nanal, Wu, and Maruyama, Nanal further discloses: The one or more non-transitory computer-readable media of claim 10, wherein the one or more computer programs are further configured to cause the at least one processor to:
determine that the AI model is not achieving an outcome … (at least as in paragraph 0045, wherein “If it is determined at 412 that no solutions exist for implementing to resolve the issue 152, then the issue is presented to the user for review at 414”); and
provide an indication to a server to retrain the AI model (at least as in paragraph 0042, wherein after the system determines no solution exists, the user may provide feedback with the correct solution for the new issue and “the user gestures 356 to fix the issue 152 can be recorded as a new solution within the knowledge base 108 and the bots for implementing the correct solution may be trained on the user gestures 356 by the bot generator 308”).
But Nanal does not specifically disclose “after a predetermined number of modifications to the confidence threshold.”
He discloses a system for classifying defects detected during the semiconductor fabrication process wherein the defect classifier utilizing machine learning is modified using a training set that includes the defects classified with a confidence below a threshold. He specifically discloses wherein an automated classifier may be created with auto confidence threshold that increases the confidence threshold from a minimum value until it reaches a target level of purity (at least as in paragraph 0053). He additionally discloses wherein the number of novel defects detected by the classifier exceeds a threshold (at least as in paragraphs 0092-0093 ). He further discloses wherein once the number of novel defect exceeds a threshold, re-training of the classifier may be triggered (at least as in paragraph 0092-0093).
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include He’s teaching of reaching a purity target with an automatic confidence threshold and retraining after a certain number of novel defects, since He teaches wherein the automatic confidence threshold and retraining improves accuracy and reduces false positives thus improving the cost of ownership and amount of manual review.
Regarding claim 17, the above combination of Nanal, Wu, and Maruyama discloses the one or more computing systems of claim 16, however, Nanal does not specifically disclose “wherein the modification of the one or more confidence threshold ranges by the automation is repeated until a winning state is achieved.”
Maruyama discloses an abnormality detection apparatus that acquires and determines a confidence or predictive value through the use of machine learning and other predictive models to detect the presence or absence of abnormality of the monitoring target apparatus. Maruyama specifically discloses wherein the abnormality detection method utilizes the upper and lower limit thresholds “UCL2” and “LCL2” in addition to the upper and lower limit thresholds “UCL1” and “LCL1” (at least as in paragraph 0126-0127), and further wherein “the abnormality detection apparatus 1 dynamically changes the upper and the lower limit thresholds” (at least as in paragraph 0129). Maruyama further discloses wherein one run of a process may be a process for a batch in a batch process or a process for a wafer in a sheet process (at least as in paragraph 0068). Maruyama additionally discloses wherein the semiconductor manufacturing apparatus transmits the observation value, or a value indicating the operating state of the step at predetermined timing of the step executed in the semiconductor manufacturing apparatus, to the observation value acquisition unit when the same process is repeated a predetermined number of times (at least as in paragraphs 0067 and 0068). Thus, Maruyama teaches “the modification of the one or more confidence threshold ranges by the automation.”
He discloses a system for classifying defects detected during the semiconductor fabrication process wherein the defect classifier utilizing machine learning is modified using a training set that includes the defects classified with a confidence below a threshold. He specifically discloses wherein an automated classifier may be created with auto confidence threshold that increases the confidence threshold from a minimum value until it reaches a target level of purity (at least as in paragraph 0053). Thus, He teaches “until a winning state is achieved.”
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include Maruyama’s teaching of dynamically changing numerous upper and lower limit thresholds and transmitting the observation value when the same process is repeated a predetermined number of times and He’s teaching of reaching a purity target with an automatic confidence threshold, since Maruyama teaches wherein the threshold ranges, dynamic adaptation, and detection transmission have an effect of achieving accurate and efficient abnormality detection and He teaches wherein the automatic confidence threshold improves the cost of ownership as the improved accuracy and reduced false positives require less manual review.
Regarding claim 18, in view of the above combination of Nanal, Wu, and Maruyama, Nanal further discloses the one or more computing systems of claim 16, further comprising: determining, by the automation, that the at least one AI models are not achieving an outcome after a predetermined number at least as in paragraph 0032 and 0020, wherein “If no prior solution can be identified, the issue can be categorized as a new issue and a further methodology such as nearest neighbor can be employed to determine matches between the prior solutions in the knowledge base and the issue,” and further wherein “If no prior solution is still found for the issue, the new issue is then presented to the user for user input”) …
However, Nanal does not specifically disclose “after a predetermined number of modifications to the confidence threshold ranges; and retraining the at least one AI models.”
He discloses a system for classifying defects detected during the semiconductor fabrication process wherein the defect classifier utilizing machine learning is modified using a training set that includes the defects classified with a confidence below a threshold. He specifically discloses wherein an automated classifier may be created with auto confidence threshold that increases the confidence threshold from a minimum value until it reaches a target level of purity (at least as in paragraph 0053). He additionally discloses wherein the number of novel defects detected by the classifier exceeds a threshold (at least as in paragraphs 0092-0093 ). He further discloses wherein once the number of novel defect exceeds a threshold, re-training of the classifier may be triggered (at least as in paragraph 0092-0093).
Therefore, it would have been obvious to one of the ordinary skill in the art at the effective filing date of the instant invention to modify the teachings of Nanal, to include He’s teaching of reaching a purity target with an automatic confidence threshold and retraining after a certain number of novel defects, since He teaches wherein the automatic confidence threshold and retraining improves accuracy and reduces false positives thus improving the cost of ownership and amount of manual review.
Claim(s) 7 and 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over S Nanal et al. (US 20200180148 A1, hereinafter Nanal) in view of Barad (US 2019/0362269 A1, hereinafter Barad).
Regarding claim 7, Nanal further discloses: The computer-implemented method of claim 1, wherein the automation calls (at least as in paragraph 0038, wherein “the solution evaluator 302 can implement a decision tree methodology for generating scores 322 for the solutions 312 based on matches between the features of the issue 152 and the features of each of the solutions”; at least as in paragraph 0039, wherein “The scores 322 are used by the issue categorizer 304 to categorize the issue 152 as one of a new issue or an existing issue”; at least as in paragraph 0050, wherein “The decision tree methodology enables identifying the correct solution for the issue arising . . . via traversing the tree paths or branches of the tree structure 700 . . . Each feature of the solution may be associated with a respective score based on a match between the solution feature and the issue features).
However, Nanal does not specifically disclose wherein “multiple … combined to determine a global confidence value that is compared.”
Barad discloses a method to self-generate a multiple-output ensemble model utilizing multiple models to defend against adversarial attacks. Barad additionally teaches wherein the output layer of the ensemble model can calculate any output type, such as a confidence score (see at least paragraph 0017), and said output data may undergo post-processing after it is generated by the AI model to transform the output into a useful result, such as aggregating the multiple outputs generated by the ensemble model (see at least paragraphs 0043-0045).
Therefore, it would have been obvious to one of ordinary skill in the art at the effective filing date of the instant invention to modify the teachings Nanal, to include Barad’s teaching of utilizing an ensemble model and aggregating the multiple confidence score outputs, since Barad teaches wherein such method enables a reduction of the variance and error rate of predictions.
Regarding claim 8, in view of the above combination of Nanal and Barad, Nanal further discloses: The computer-implemented method of claim 7, wherein the global confidence value is determined by applying a respective weight to the confidence values and combining the weighted confidence values.
Barad discloses a method to self-generate a multiple-output ensemble model utilizing multiple models to defend against adversarial attacks. Barad specifically discloses wherein the ensemble model’s output values, such as confidence scores, undergo post-processing such as aggregating into a single output using weighted averaging (at least as in paragraphs 0020 and 0043).
Therefore, it would have been obvious to one of ordinary skill in the art at the effective filing date of the instant invention to modify the teachings Nanal, to include Barad’s teaching of utilizing an ensemble model and aggregating the multiple confidence score outputs, since Barad teaches wherein such method enables a reduction of the variance and error rate of predictions.
Claim(s) 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over S Nanal et al. (US 20200180148 A1, hereinafter Nanal) in view of Wu et al (US 2019/0205761 A1, hereinafter Wu) and Maruyama (US 20200333777 A1), and further in view of Barad (US 2019/0362269 A1, hereinafter Barad).
The teachings of Nanal, Wu, and Maruyama have been discussed above.
Regarding claim 20, the above combination of Nanal, Wu, and Maruyama teaches the computer-implemented method of claim 14, however, Nanal does not specifically disclose wherein the RPA robot calls multiple AI models and the confidence values from each AI model are combined to determine a global confidence value that is compared against the confidence threshold ranges for the probabilistic activity, and the global confidence value is determined by applying a respective weight to the confidence values and combining the weighted confidence values.
Barad discloses a method to self-generate a multiple-output ensemble model utilizing multiple models to defend against adversarial attacks. Barad additionally teaches wherein the output layer of the ensemble model can calculate any output type, such as a confidence score (see at least paragraph 0017), and said output data may undergo post-processing after it is generated by the AI model to transform the output into a useful result, such as aggregating the multiple outputs generated by the ensemble model into a single output using weighted averaging (see at least paragraphs 0020 and 0043-0045). Barad further discloses wherein the ensemble model’s may calculate any output value such as confidence scores (at least as in paragraph 0017).
Therefore, it would have been obvious to one of ordinary skill in the art at the effective filing date of the instant invention to modify the teachings Nanal, to include Barad’s teaching of utilizing an ensemble model and aggregating the multiple confidence score outputs, since Barad teaches wherein such method enables a reduction of the variance and error rate of predictions.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RICARDO ICHIKAWA VISCARRA whose telephone number is (571)270-0154. The examiner can normally be reached M-F 12:00 - 20:00 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, Adam Mott can be reached on (571) 270-5376. 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.
/RICARDO I VISCARRA/Examiner, Art Unit 3657
/ADAM R MOTT/Supervisory Patent Examiner, Art Unit 3657
Prosecution Insights