DETAILED ACTION
This action is responsive to the application filed on August 23, 2024, which is continuation of 18/093,874 filed on January 06, 2023, now US Pat. No. 12,099,435.
Claims 1-9 are pending and presented for examination.
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Examiner Notes
Examiner cites particular columns, paragraphs, figures and line numbers in the references as applied to the claims below for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested that, in preparing responses, the applicant fully consider the references in their entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the examiner.
Drawings
The drawings filed on August 23, 2024 are acceptable for examination purposes.
Information Disclosure Statement
As required by M.P.E.P. 609, the applicant’s submission of the Information Disclosure Statement dated September 16, 2024 is acknowledged by the examiner and the cited references have been considered in the examination of the claims now pending.
Specification
The disclosure is objected to because of the following informalities: The specification does not contain the CROSS-REFERENCE TO RELATED APPLICATIONS area. In addition, paragraph [0002] of the specification recites “by mimicking mimic human-like behavior,” which appears to contain a duplicated or extraneous term. Appropriate correction is required.
Claim Objections
Claims 1-4 are objected to because of the following informalities: each of claims 1 and 3 recites “simulates application of the potential solution; and” followed by further limitations, such that the conjunction “and” appears in the middle of the recited list of limitations rather than before the final limitation of the list. Claims 2 and 4 are objected to as depending from an objected-to claim. 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 1-9 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1-13 of U.S. Pat. No. 12,099,435. Although the instant claims and the claims of U.S. Patent No. 12,099,435 are not identical, they are not patentably distinct because the instant claims recite substantially the same subject matter as, and define a scope encompassed by, the corresponding patent claims, as set forth below.
Instant claim 1 recites the same method as claim 5 of the patent and differs only in that instant claim 1 conditions the adjustment of the at least one processing parameter on a timestamp associated with each of the inputs, whereas claim 5 of the patent conditions the adjustment on a geographic location associated with the inputs. A timestamp and a geographic location are each a readily available attribute of an input, and this difference does not patentably distinguish instant claim 1 from claim 5 of the patent.
Instant claim 3 recites the same method as claim 5 of the patent and differs only in that the action taken in response to determining that the aberrant outputs are due to aberrant processing of the inputs is to decommission the bot, rather than to adjust the at least one processing parameter. Decommissioning a bot that has been determined to be malfunctioning is an obvious alternative remedial response, and this difference does not patentably distinguish instant claim 3 from claim 5 of the patent.
Instant claim 5 recites a system the scope of which is encompassed by claims 1, 2, and 4 of the patent. Claim 4 of the patent recites that the secondary bot is a digital twin of the bot; claim 1 of the patent recites tracing the processing of the secondary bot, generating and submitting test inputs, detecting the threshold number of aberrant outputs generated by the bot, and formulating and applying a remedial action based on the results of the test inputs before applying it; and claim 2 of the patent recites that the remedial action reprograms the bot to change a processing parameter. Instant claim 5 is therefore not patentably distinct from claims 1, 2, and 4 of the patent.
Instant claims 2 and 4 recite limitations that are recited verbatim in claim 8 of the patent, and instant claims 6, 7, 8, and 9 recite limitations that are recited verbatim, respectively, in claims 10, 11, 12, and 13 of the patent. These dependent claims are therefore not patentably distinct from the corresponding claims of the patent.
Instant Application
U.S. Pat. No. 12,099,435
Claim 1. An artificial intelligence (“AI”) method for autonomously diagnosing a malfunction of a bot … wherein execution of the computer readable instructions by the processor: monitors outputs generated by the bot; detects a threshold number of aberrant outputs during a time-window; identifies a potential solution for reducing the threshold number of aberrant outputs; simulates application of the potential solution; based on results of the simulating, autonomously adjusts at least one processing parameter of the bot; parses inputs associated with each of the aberrant outputs; determines whether the threshold number of aberrant outputs are due to aberrant inputs or aberrant processing of the inputs by the bot; and in response to determining that the threshold number of aberrant outputs are due to aberrant processing of the inputs by the bot, adjusts the at least one processing parameter of the bot based on a timestamp associated with each of the inputs.
Claim 5. An artificial intelligence (“AI”) method for autonomously diagnosing a malfunction of a bot … wherein execution of the computer readable instructions by the processor: monitors outputs generated by the bot; detects a threshold number of aberrant outputs during a time-window; identifies a potential solution for reducing the threshold number of aberrant outputs; simulates application of the potential solution; based on results of the simulating, autonomously adjusts at least one processing parameter of the bot; parses inputs associated with each of the aberrant outputs; determines whether the threshold number of aberrant outputs are due to aberrant inputs or aberrant processing of the inputs by the bot; and in response to determining that the threshold number of aberrant outputs are due to aberrant processing of the inputs by the bot, adjusts the at least one processing parameter of the bot based on a geographic location associated with inputs received by the bot after the time-window.
Claim 3. An artificial intelligence (“AI”) method for autonomously diagnosing a malfunction of a bot … [reciting the limitations recited opposite instant claim 1, supra, through] determines whether the threshold number of aberrant outputs are due to aberrant inputs or aberrant processing of the inputs by the bot; and in response to determining that the threshold number of aberrant outputs are due to aberrant processing of the inputs by the bot, decommissions the bot.
Claim 5. [the method limitations recited opposite instant claim 1, supra] … in response to determining that the threshold number of aberrant outputs are due to aberrant processing of the inputs by the bot, adjusts the at least one processing parameter of the bot based on a geographic location associated with inputs received by the bot after the time-window.
Claim 5. An artificial intelligence (“AI”) system for autonomously diagnosing a malfunction with a bot … a first bot that is programmed to receive user inputs and generate automated outputs; a second bot that is programmed to: monitor the user inputs and automated outputs generated by the first bot during a first time-window; detect a threshold number of aberrant automated outputs generated by the first bot; diagnose a malfunction that is causing the first bot to generate the threshold number of aberrant automated outputs; autonomously reprogram the first bot to change a processing parameter applied by the first bot; wherein the second bot is further programmed to simulate the change to the processing parameter before autonomously reprogramming the first bot; generate a test input; input the test input to a digital twin of the first bot; and trace each operational step performed by the digital twin to generate an automated response in response to the test input.
Claim 1. An artificial intelligence (“AI”) system for autonomously diagnosing and remediating a malfunction of a bot … a supervisory bot programmed to: during a first time-window, monitor automated outputs generated by the plurality of bots and detect a target bot generating a threshold number of aberrant automated outputs; and during a second time-window, input a user input to a secondary bot; trace a processing by the secondary bot of the user input; in response to detecting a divergence in the processing of the user input by the target bot and the secondary bot, generate a plurality of test inputs; submit the plurality of test inputs to the target bot and the secondary bot; formulate a remedial action based on the resulting automated outputs; and apply the remedial action to the target bot.
Claim 2. … the remedial action comprises reprogramming the target bot to change a processing parameter applied by the target bot.
Claim 4. … the secondary bot is a digital twin of the target bot; and the digital twin is generated prior to a start of the first time-window.
Claim 2
Claim 8
Claim 4
Claim 8
Claim 6
Claim 10
Claim 7
Claim 11
Claim 8
Claim 12
Claim 9
Claim 13
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 and 3 are rejected under 35 U.S.C. 103 as being unpatentable over Higgins et al. (US Pub. No. 2023/0122872, hereinafter Higgins) in view of Johnson et al. (US Pat. No. 10,652,139, hereinafter Johnson) in view of Ganesan et al. (US Pat. No. 11,726,902, hereinafter Ganesan) and further in view of Menon et al. (US Pub. No. 2023/0188480, hereinafter Menon – IDS 09/16/2024).
With respect to claim 1, Higgins teaches an artificial intelligence ("AI") method for autonomously diagnosing a malfunction of a bot, the method comprising extracting computer readable instructions stored on a non-transitory medium and executing the computer readable instructions on a processor, wherein execution of the computer readable instructions by the processor: (Higgins discloses a computer-implemented method and system that provides a framework for real-time identification of bot conversation issues and for evaluating and improving the performance of an automated conversation bot agent (Higgins, Abstract; paragraphs [0002], [0005]). The method is carried out by a system comprising one or more processors and memory storing executable instructions, and by a non-transitory computer-readable storage medium storing instructions that, when executed by the one or more processors, cause the recited processes to be performed (Higgins, paragraph [0004]); a bot agent is code that, when executed by the processor, autonomously communicates with users (Higgins, paragraph [0028]). Higgins therefore teaches an AI method, embodied in computer-readable instructions on a non-transitory medium executed by a processor, for autonomously diagnosing a malfunction (degraded performance) of a bot). monitors outputs generated by the bot (Higgins records and monitors the conversation between a conversation bot agent and a customer, including the messages (outputs) generated by the bot agent, and the bot state machine learning modeling engine processes those messages in real time to evaluate the bot agent’s performance (Higgins, paragraphs [0003], [0031], [0043], [0045])). detects [[a threshold number of]] aberrant outputs [[during a time-window]] (Higgins detects a set of bot states corresponding to the performance of the bot agent, including failure states such as the bot failing to understand the customer’s intent, ignoring the customer, becoming stuck in a message loop, customer frustration, erroneous transfers, reprompts, and fallbacks in which the bot asks the customer to rephrase—i.e., aberrant outputs of the bot (Higgins, paragraphs [0031], [0033], [0049])).
parses inputs associated with each of the aberrant outputs (Higgins parses the user inputs (customer messages) associated with each detected aberrant bot state, extracting features from those messages—such as reprompts, number of fallbacks, unplanned escalations, and repeated intents—on a per-message basis for the messages tied to each detected state (Higgins, paragraphs [0031], [0049], [0050])). Higgins detects the aberrant outputs but does not expressly quantify that detection as a threshold number occurring during a time-window, however in an analogous art, Johnson teaches a threshold number of aberrant outputs during a time-window (Johnson teaches a monitoring system that analyzes records from a time slice corresponding to a particular time window—for example, the last ten minutes—and detects an error condition when the number of anomalous events within that window exceeds an expected number by a threshold (Johnson, column 2 lines 40-64, column 4 lines 4-20, column 4 lines 37-43, column 8 lines 6-33). Applying Johnson’s threshold-within-a-time-window detection to Higgins’ detection of aberrant bot outputs teaches detecting a threshold number of aberrant outputs during a time-window)).
It would have been obvious to one of ordinary skill in the art at the time the invention was made before the effective filing date of the claimed invention to apply Johnson’s threshold-within-a-time-window detection to Higgins’ monitoring of aberrant bot outputs, in order to provide an objective, automated trigger for determining when a bot’s aberrant outputs are sufficiently frequent and persistent to warrant remediation, thereby detecting difficult-to-detect, persistent errors and minimizing service interruptions—the express benefit taught by Johnson (Johnson, column 1 line 62 – column 2 line 25, column 2 line 65 – column 3 line 29). The combination is no more than the use of a known monitoring technique to improve an analogous bot-monitoring system in the same way, yielding predictable results.
Higgins detects and diagnoses the aberrant bot states but relies on human bot builders to devise and apply the corresponding fix (Higgins, paragraphs [0003], [0044]). Higgins does not disclose the processor itself identifying, simulating, or attributing the cause of the aberrant outputs. Higgins in view of Johnson is silent to disclose, however in an analogous art, Ganesan teaches identifies a potential solution for reducing the threshold number of aberrant outputs (Ganesan’s scenario simulator operates as an optimization engine that identifies and suggests improvements and corrections for the bot’s functionality, and its coding recommender generates an execution plan setting forth recommendations for improving or correcting the bot, including a specific coding change to be made (Ganesan, column 7 lines 30-50, column 8 lines 10-32)). simulates application of the potential solution (Ganesan creates a plurality of inputs for simulation of the bot’s functionality and executes the bot against those simulated inputs to test and validate the bot, including the corrected bot, before deployment (Ganesan, Abstract; column 7 lines 30-50, column 8 lines 33-49, column 10 lines 6-18)).
determines whether the threshold number of aberrant outputs are due to aberrant inputs or aberrant processing of the inputs by the bot (Ganesan automatically validates the bot to determine whether the bot is defective: it applies benchmarking logic and generates a score, where a score that fails to reach a predetermined threshold (or a failure result returned from any module) indicates a defective bot—i.e., the bot’s processing is at fault—whereas a non-defective bot is deployed (Ganesan, column 8 lines 33-49, column 10 lines 6-18). Separately, Ganesan’s data anomaly simulator executes the bot against different, anomalous input samples to test whether the anomalous inputs, rather than the bot, produce the exceptional conditions (Ganesan, column 7 line 51 – column 8 line 9). Ganesan thereby determines whether the aberrant outputs are due to aberrant inputs or to aberrant processing by the bot).
It would have been obvious to one of ordinary skill in the art at the time the invention was made before the effective filing date of the claimed invention to incorporate Ganesan’s automated identification of a corrective recommendation, simulation/testing, and defective-bot determination into the system of Higgins/Johnson, in order to automatically identify and validate a proposed fix before it is applied and to distinguish a genuinely defective bot from anomalous inputs, thereby avoiding the deployment of ineffective fixes and strengthening the bot before redeployment—the express benefit taught by Ganesan (Ganesan, column 3 lines 34-47, column 8 lines 50-59). This is the predictable use of a known automated bot-testing technique to improve an analogous bot system in the same way.
Higgins implements the bot update through human bot builders and therefore does not autonomously adjust the bot.
Higgins in view of Johnson in view of Ganesan is silent to disclose, however in an analogous art, Menon teaches based on results of the simulating, autonomously adjusts at least one processing parameter of the bot (Menon teaches a reinforcement-learning chatbot that is trained and corrected through trial-and-error conversations with user simulators and that, based on the results of that simulation/trial, autonomously adjusts at least one processing parameter of the bot: upon detecting a negative-reward (aberrant) response, Menon’s RL engine automatically retrains the RL model(s) and adjusts the response-generation parameters—for example, by modifying the weights associated with recommendations—to produce a corrected response, expressly automating the detection, retraining, and adjustment steps that prior systems performed manually (Menon, Abstract; paragraphs [0004], [0026], [0028]-[0029], [0064], [0075])).
in response to determining that the threshold number of aberrant outputs are due to aberrant processing of the inputs by the bot, adjusts the at least one processing parameter of the bot [[based on a timestamp associated with each of the inputs]] (Menon teaches that, in response to determining that the bot’s processing is at fault — i.e., that its response is assigned a negative reward — the system adjusts the at least one processing parameter of the bot by automatically retraining the RL model(s) and modifying the response-generation parameters to produce a corrected response (Menon, paragraphs [0004],[0028]-[0029], [0064], [0075]). Higgins in view of Ganesan in view of Menon is silent to disclose; however, in an analogous art, Johnson teaches adjusting based on a timestamp associated with each of the inputs (Johnson identifies the records to be acted upon by the time associated with each record, obtaining a time slice of records corresponding to a particular time window—for example, the communications occurring within the last ten minutes or the last half-hour—and, upon determining that the error condition is present within that time slice, acts to correct or avoid the error condition affecting that traffic (Johnson, column 4 lines 37-43, column 8 lines 6-33). The correction in Johnson is therefore made based on the time associated with each record/input). It would have been obvious to one of ordinary skill in the art at the time the invention was made before the effective filing date of the claimed invention to base the adjustment of the bot’s processing parameter on a timestamp associated with each of the inputs, as taught by Johnson, in order to target the correction to the particular time-window in which the aberrant outputs occur, thereby recalibrating the bot for the traffic actually affected and minimizing service interruptions, with a reasonable expectation of success because Johnson’s time-based error detection is already incorporated into the combination as set forth above.) It would have been obvious to one of ordinary skill in the art at the time the invention was made before the effective filing date of the claimed invention to incorporate Menon’s autonomous reinforcement-learning self-correction into the system of Higgins/Johnson/Ganesan, so that the bot is adjusted autonomously rather than through human bot builders, because Menon teaches that automating the detection of feedback, retraining, and response adjustment—steps that existing systems perform manually—improves the accuracy of responses, increases the reliability of the system, and reduces the computing and network resources otherwise expended (Menon, paragraphs [0028], [0075], [0085]). This is the predictable substitution of an autonomous correction mechanism for a manual one to accomplish the same purpose.
With respect to claim 3, claim 3 recites an artificial intelligence method identical to the method of claim 1 except that, in place of adjusting the at least one processing parameter based on a timestamp, claim 3 recites that, in response to determining that the threshold number of aberrant outputs are due to aberrant processing of the inputs by the bot, the processor decommissions the bot. The preamble and the limitations common to claims 1 and 3—monitoring outputs; detecting a threshold number of aberrant outputs during a time-window; identifying a potential solution; simulating application of the potential solution; based on results of the simulating, autonomously adjusting at least one processing parameter; parsing inputs associated with each aberrant output; and determining whether the aberrant outputs are due to aberrant inputs or aberrant processing—are rejected for the same reasons, and over the same references, set forth above in the rejection of claim 1. As to the differing limitation, Ganesan teaches in response to determining that the threshold number of aberrant outputs are due to aberrant processing of the inputs by the bot, decommissions the bot (Ganesan teaches that, in response to determining that the bot is defective (i.e., that the bot’s own processing is at fault), the bot is not deployed but is instead taken out of service and returned for redevelopment (Ganesan, column 8 lines 33-59, column 10 lines 6-18)—thereby decommissioning the bot. The motivation to combine Ganesan with Higgins, Johnson, and Menon is the same as set forth in the rejection of claim 1).
Claims 2 and 4 are rejected under 35 U.S.C. 103 as being unpatentable over Higgins et al. (US Pub. No. 2023/0122872, hereinafter Higgins) in view of Johnson et al. (US Pat. No. 10,652,139, hereinafter Johnson) in view of Ganesan et al. (US Pat. No. 11,726,902, hereinafter Ganesan) in view of Menon et al. (US Pub. No. 2023/0188480, hereinafter Menon – IDS 09/16/2024) and further in view of Radhakrishnan et al. (US Pub. No. 2020/0065223 – IDS 09/16/2024). With respect to claim 2, Higgins in view of Johnson in view of Menon is silent to disclose, however in an analogous art, Ganesan teaches: generates simulated inputs (Ganesan creates a plurality of inputs for simulation of the detected functionality of the bot, drawn from scenario libraries associated with that functionality (Ganesan, Abstract; column 7 lines 30-50, column 9 lines 34-42)). submits the simulated inputs to the bot (Ganesan executes the input bot a plurality of times using the sample(s) of actions derived from the created inputs (Ganesan, Abstract; column 7 line 51 – column 8 line 9, column 9 lines 43-65)). registers the bot as being associated with a processing error when the outputs generated by the bot in response to the simulated inputs [[do not correspond to the known outputs]] (Ganesan deems the bot defective—and returns the defective bot for redevelopment—when the score fails to reach the predetermined threshold or when any module returns a failure result, thereby registering the bot as being associated with a processing error (Ganesan, column 8 lines 33-59, column 10 lines 6-18)). registers the inputs as being erroneous when the outputs generated by the bot in response to the simulated inputs [[correspond to the known outputs]] (When the bot is non-defective, Ganesan attributes the anomalous behavior to the input data rather than the bot: its data anomaly simulator tests the exceptional conditions arising from the different, anomalous input samples, and the non-defective bot is deployed (Ganesan, column 7 line 51 – column 8 line 9, column 8 lines 50-59), thereby registering the inputs as being erroneous). Higgins in view of Johnson in view of Ganesan in view of Menon is silent to disclose, however in an analogous art, Radhakrishnan teaches generates known outputs associated with the simulated inputs (Radhakrishnan reproduces an operation in parallel to generate a simulated operation having a corresponding expected (known) result against which the operation is evaluated (Radhakrishnan, paragraphs [0008], [0010])). determines whether outputs generated by the bot in response to the simulated inputs correspond to the known outputs (Radhakrishnan compares a result of the operation to the expected (known) result and, based on the result not matching the expected result, generates an indication of the error (Radhakrishnan, paragraph [0010])). It would have been obvious to one of ordinary skill in the art at the time the invention was made before the effective filing date of the claimed invention to incorporate Radhakrishnan’s generation of an expected (known) result and its comparison of the bot’s output to that expected result into the automated bot-testing of Ganesan, and the system of Higgins/Johnson/Menon in order to provide an objective, automated criterion for determining when the bot’s output diverges from the intended output and thereby to efficiently diagnose and rectify the error—the express purpose taught by Radhakrishnan (Radhakrishnan, paragraphs [0002], [0008]). This is the predictable use of a known expected-result comparison to improve an analogous automated bot-testing system in the same way. With respect to claim 4, claim 4 depends from claim 3 and recites limitations identical to those of claim 2, and is rejected for the same reasons set forth above in the rejection of claim 2.
Allowable Subject Matter
Claim 5 is free of the prior art of record. The prior art of record, taken alone or in combination, does not teach or reasonably suggest a system that, in diagnosing a malfunction of a first bot, generates a test input, inputs the test input to a digital twin of the first bot, and traces each operational step of the digital twin to localize the malfunction, in combination with simulating the change to the first bot before autonomously reprogramming it, as recited in claim 5. While the prior art teaches monitoring a bot, detecting a threshold number of aberrant outputs, and autonomously correcting a bot (see the rejection of claims 1–4 above), and while digital twins are known generally in industrial and Internet-of-Things contexts, the art of record does not disclose or suggest generating a digital twin of the particular bot under diagnosis and tracing each of its individual operational steps to identify the source of the malfunction in the manner claimed. Claims 6–9, which depend from claim 5, are free of the prior art of record for at least the same reason. Claims 5-9 nonetheless remain subject to the nonstatutory double patenting rejection over U.S. Patent No. 12,099,435 set forth above, which may be overcome by the filing of an appropriate terminal disclaimer.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Braden Hancock et al. (“Learning from Dialogue after Deployment: Feed Yourself, Chatbot!”) The majority of conversations a dialogue agent sees over its lifetime occur after it has already been trained and deployed, leaving a vast store of potential training signal untapped. In this work, we propose the self-feeding chatbot, a dialogue agent with the ability to extract new training examples from the conversations it participates in. As our agent engages in con versation, it also estimates user satisfaction in its responses. When the conversation appears to be going well, the user’s responses become new training examples to imitate. When the agent believes it has made a mistake, it asks for feedback; learning to predict the feedback that will be given improves the chatbot’s dialogue abilities further. On the PERSONACHAT chit chat dataset with over 131k training examples, we find that learning from dialogue with a self-feeding chatbot significantly improves performance, regardless of the amount of traditional supervision. (see abstract).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANIBAL RIVERACRUZ whose telephone number is (571)270-1200. The examiner can normally be reached Monday-Friday 9:30 AM-6:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Hyung S Sough can be reached at 5712726799. 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.
/ANIBAL RIVERACRUZ/Primary Examiner, Art Unit 2192
Prosecution Insights