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 .
Compact Prosecution
Examiner would like to propose an initiated Telephone interview to discuss the merits of the case so that a proposed amendments that can move forward and prosecuting the case compactly will be achieve.
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.
Claims 1-30 are rejected under 35 U.S.C. 103 as being unpatentable over Hettige (US20250094821) in view of Kar (US20200160178)
Claim 1, Hettige discloses a method implemented by a virtual assistant server comprising:
receiving a plurality of data inputs from a developer device to create conversation experiments for a use case; (Section 0191, lines 1-5 the dialog script, such as “You are an agent to help decide which action to take based on the user question and context information” Also the prompt and response template reads on the conversation experiments)
creating each of the conversation experiments upon generating a first user utterance of each of the conversation experiments corresponding to the use case based on the plurality of data inputs, (Section 0191- thus prompts are generated using synthesized training examples- Thus the Initial User Utterance are generated for each conversational scenario)
wherein for each of the conversation experiments the creating further comprises:
generating one or more virtual assistant responses, virtual assistant actions, or subsequent user utterances, based on at least one of the plurality of data inputs or a prior version of a current one of the conversation experiments; (Section 0196, lines 3-6- thus responses are generated based on the dialog script and using corresponding response template for the predefined scenario).
(Also see Section 0160, lines 1-4 thus one or more responses are generated for the one or more prompts)
validating each of the one or more generated virtual assistant responses, the virtual assistant actions, or the subsequent user utterances against validation data, (Section 0169, lines 4-5- thus 10%-15% validation, also the Validation Phase in section 0131 reads on validating each of the validation data).
wherein the current one of the conversation experiments is accepted only upon the successful validation of the corresponding one of the one or more virtual assistant responses, the virtual assistant actions, or the subsequent user utterances; (Only validated (acceptable) synthetic conversation examples are added to the training dataset ([0168], [0198], [0131]).
and repeating the generating and the validating until one or more exit conditions are satisfied for the current one of the conversation experiments; (The process repeats the generation and validation steps until the desired amount or quality of data (exit conditions) is achieved ([0198], [0166], [0168])
and training a virtual assistant model with at least one of the created conversation experiments when the one or more exit conditions are satisfied for the at least one of the conversation experiments. (Once the synthetic and/or real conversation experiments are validated and reach the required conditions, they are used to fine-tune/train the assistant model ([0200], [0182]-[0185]).
Hettige does not disclose wherein the current one of the conversation experiments is updated only upon the successful validation of the corresponding one of the one or more virtual assistant responses,
Kar disclose wherein the current one of the conversation experiments is updated only upon the successful validation of the corresponding one of the one or more virtual assistant responses, (See sections 0032 and 0036-0047: thus The synthetic datasets are validated/improved by comparing against real-world validation sets and using feedback (distribution matching, task validation, see [0032], [0036]-[0047]). Only data that meets validation criteria (distributional similarity, task performance) is retained or used for further training)
Therefore it would have been obvious to one of ordinary skill in the art before the
effective filling date of the claimed invention to include the teaching of updating the model only when the validation is complete. The motivation is that it make the model to be improved and also handle new inputs.
Claim 2, Hettige in view of Kar (Section 0022- thus grammars reads on rules) discloses wherein the plurality of data inputs comprise: a domain name, a use case name, a use case description, use case attributes, one or more business rules, one or more conversation rules, the one or more exit conditions, details of function calls, one or more sample conversations, a summary of the one or more sample conversations, or one or more conversation templates. (Hettige: Section 0191 Inputs include domain name, use case, business rules, templates, etc. are used as inputs into the virtual assistant- Also see Section 0143, lines 9-12- simple rules-based systems are used as data inputs)
Claim 3, Hettige in view of Kar (Section 0022 and 0023) discloses wherein the plurality of data inputs received from the user device is in natural language text. (Section 0143, lines 2 and 6 users through “natural language” and “user input” these are natural text)
Claim 4, Hettige in view of Kar discloses wherein the plurality of data inputs received from the user device is in a structured data format. (Hettige: Section 0152: Templates and data are in structured formats (e.g., JSON)) (Also see 0144)
Claim 5, Hettige in view of Kar (Section 0022-0023) discloses wherein the structured data format is a JavaScript Object Notation (JSON) format. (Section 0153, lines 16 “args JSON schema”)
Claim 6, Hettige in view of Kar (Section 0023) discloses wherein the validation data comprises one or more of: the prior version of the current one of the conversation experiments, one or more business rules, one or more conversation rules, the one or more exit conditions, use case attributes, one or more conversation templates, or user sentiment. (Hettige: Sections 0168: Validation uses these prompt and response template See Section 0033, lines 10-15)
Claim 7, Hettige in view of Kar discloses further comprising, prior to the repeating, generating a reason for validation failure when at least one of the one or more subsequent user utterances, the one or more virtual assistant responses, or the one or more virtual assistant actions fails the validation. (Hettige: Section 0131, Validation process includes error analysis and feedback Also see Section 0168]).
Claim 8, Hettige in view of Kar (task performance scores) as feedback for updating the generative model ([0036]-[0047]). discloses wherein the reason for validation failure is used as feedback for generating: one or more subsequent virtual assistant responses, one or more subsequent virtual assistant actions, or the one or more subsequent user utterances, to avoid any further validation failures. (Hettige: Section 0168, the feedback is used to improve synthetic data generation by retraining the neural network)
Claim 9, Hettige in view of Kar discloses wherein the one or more exit conditions comprise: a user utterance or a virtual assistant response comprising one or more keywords indicating completion of a conversation experiment, (Hettige: Section 0198, lines 5-6 …”Generation of synthetic datapoints of a desired size reads on the indication of keywords completion…”) or a validation failure has occurred successively for a threshold number of times for the current one of the conversation experiments. (Hettige: Iterative process stops on exit conditions [0168]).
Claim 10, Hettige in view of Kar discloses wherein the generating and the validating are performed by prompting one or more language models. (Hettige: Sections 0168: Validation uses these prompt and response template See Section 0033, lines 10-15)
Claim 11, Hettige discloses a virtual assistant server comprising
one or more processors; (Processing Unit 1004 shown in Fig. 10) and
a memory coupled to the one or more processors which are configured to execute programmed instructions stored in the memory to: (System memory 1010 shown in Fig. 10)
receive a plurality of data inputs from a developer device to create conversation experiments for a use case; (Section 0191, lines 1-5 the dialog script, such as “You are an agent to help decide which action to take based on the user question and context information” Also the prompt and response template reads on the conversation experiments)
create each of the conversation experiments upon generating a first user utterance of each of the conversation experiments corresponding to the use case based on the plurality of data inputs, (Section 0191- thus prompts are generated using synthesized training examples- Thus the Initial User Utterance are generated for each conversational scenario)
wherein for each of the conversation experiments the creating further comprises:
generating one or more virtual assistant responses, virtual assistant actions, or subsequent user utterances, based on at least one of the plurality of data inputs or a prior version of a current one of the conversation experiments; (Section 0196, lines 3-6- thus responses are generated based on the dialog script and using corresponding response template for the predefined scenario).
(Also see Section 0160, lines 1-4 thus one or more responses are generated for the one or more prompts)
validating each of the one or more generated virtual assistant responses, the virtual assistant actions, or the subsequent user utterances against validation data,
(Section 0169, lines 4-5- thus 10%-15% validation, also the Validation Phase in section 0131 reads on validating each of the validation data).
wherein the current one of the conversation experiments is updated only upon the successful validation of the corresponding one of the one or more virtual assistant responses, the virtual assistant actions, or the subsequent user utterances; (Only validated (acceptable) synthetic conversation examples are added to the training dataset ([0168], [0198], [0131]).
and
repeating the generating and the validating until one or more exit conditions are satisfied for the current one of the conversation experiments; (The process repeats the generation and validation steps until the desired amount or quality of data (exit conditions) is achieved ([0198], [0166], [0168])
and train a virtual assistant model with at least one of the created conversation experiments when the one or more exit conditions are satisfied for the at least one of the conversation experiments. (Once the synthetic and/or real conversation experiments are validated and reach the required conditions, they are used to fine-tune/train the assistant model ([0200], [0182]-[0185]).
Hettige does not disclose wherein the current one of the conversation experiments is updated only upon the successful validation of the corresponding one of the one or more virtual assistant responses,
Kar disclose wherein the current one of the conversation experiments is updated only upon the successful validation of the corresponding one of the one or more virtual assistant responses, (See sections 0032 and 0036-0047: thus The synthetic datasets are validated/improved by comparing against real-world validation sets and using feedback (distribution matching, task validation, see [0032], [0036]-[0047]). Only data that meets validation criteria (distributional similarity, task performance) is retained or used for further training)
Therefore it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to include the teaching of updating the model only when the validation is complete. The motivation is that it make the model to be improved and also handle new inputs.
Claim 12, Hettige in view of Kar (Section 0022- thus grammars reads on rules) discloses wherein the plurality of data inputs comprise: a domain name, a use case name, a use case description, use case attributes, one or more business rules, one or more conversation rules, the one or more exit conditions, details of function calls, one or more sample conversations, a summary of the one or more sample conversations, or one or more conversation templates. (Hettige: Section 0191 Inputs include domain name, use case, business rules, templates, etc. are used as inputs into the virtual assistant- Also see Section 0143, lines 9-12- simple rules-based systems are used as data inputs)
Claim 13, Hettige in view of Kar (Section 0022 and 0023) discloses wherein the plurality of data inputs received from the user device is in natural language text. (Section 0143, lines 2 and 6 users through “natural language” and “user input” these are natural text)
Claim 14, Hettige in view of Kar discloses wherein the plurality of data inputs received from the user device is in a structured data format. (Hettige: Section 0152: Templates and data are in structured formats (e.g., JSON)) (Also see 0144)
Claim 15, Hettige in view of Kar (Section 0022-0023) discloses wherein the structured data format is a JavaScript Object Notation (JSON) format. (Section 0153, lines 16 “args JSON schema”)
Claim 16, Hettige in view of Kar discloses wherein the validation data comprises one or more of the prior version of the current one of the conversation experiments, one or more business rules, one or more conversation rules, the one or more exit conditions, use case attributes, one or more conversation templates, or user sentiment. (Hettige: Sections 0168: Validation uses these prompt and response template See Section 0033, lines 10-15)
Claim 17, Hettige in view of Kar (task performance scores) as feedback for updating the generative model ([0036]-[0047]) discloses wherein prior to the repeating, the one or more processors are further configured to execute the programmed instructions stored in the memory to: generate a reason for validation failure when at least one of the one or more subsequent user utterances, the one or more virtual assistant responses, or the one or more virtual assistant actions fails the (Hettige: Section 0131, Validation process includes error analysis and feedback Also see Section 0168]).
Claim 18, Hettige in view of Kar discloses wherein the reason for validation failure is used as feedback to generate: one or more subsequent virtual assistant responses, one or more subsequent virtual assistant actions, or the one or more subsequent user utterances, to avoid any further validation failures. (Hettige: Section 0168, the feedback is used to improve synthetic data generation by retraining the neural network)
Claim 19, Hettige in view of Kar discloses wherein the one or more exit conditions comprise: a user utterance or a virtual assistant response comprising one or more keywords indicating completion of a conversation experiment, (Hettige: Section 0198, lines 5-6 …”Generation of synthetic datapoints of a desired size reads on the indication of keywords completion…”) or a validation failure has occurred successively for a threshold number of times for the current one of the conversation experiments. (Hettige: Iterative process stops on exit conditions [0168]).
Claim 20, Hettige in view of Kar discloses wherein the generating and the validating are performed by prompting one or more language models. (Hettige: Sections 0168: Validation uses these prompt and response template See Section 0033, lines 10-15)
Claim 21, Hettige discloses a non-transitory computer-readable medium storing instructions which when executed by one or more processors, (Processing Unit 1004 shown in Fig. 10) causes the one or more processors to:
receive a plurality of data inputs from a developer device to create conversation experiments for a use case; (Section 0191, lines 1-5 the dialog script, such as “You are an agent to help decide which action to take based on the user question and context information” Also the prompt and response template reads on the conversation experiments)
create each of the conversation experiments upon generating a first user utterance of each of the conversation experiments corresponding to the use case based on the plurality of data inputs, (Section 0191- thus prompts are generated using synthesized training examples- Thus the Initial User Utterance are generated for each conversational scenario)
wherein for each of the conversation experiments the creating further comprises:
generating one or more virtual assistant responses, virtual assistant actions, or subsequent user utterances, based on at least one of the plurality of data inputs or a prior version of a current one of the conversation experiments; (Section 0196, lines 3-6- thus responses are generated based on the dialog script and using corresponding response template for the predefined scenario).
(Also see Section 0160, lines 1-4 thus one or more responses are generated for the one or more prompts)
validating each of the one or more generated virtual assistant responses, the virtual assistant actions, or the subsequent user utterances against validation data, (Section 0169, lines 4-5- thus 10%-15% validation, also the Validation Phase in section 0131 reads on validating each of the validation data)
wherein the current one of the conversation experiments is accepted only upon the successful validation of the corresponding one of the one or more virtual assistant responses, the virtual assistant actions, or the subsequent user utterances; (Only validated (acceptable) synthetic conversation examples are added to the training dataset ([0168], [0198], [0131]).
and repeating the generating and the validating until one or more exit conditions are satisfied for the current one of the conversation experiments; (The process repeats the generation and validation steps until the desired amount or quality of data (exit conditions) is achieved ([0198], [0166], [0168])
and train a virtual assistant model with at least one of the created conversation experiments when the one or more exit conditions are satisfied for the at least one of the conversation experiments. (Once the synthetic and/or real conversation experiments are validated and reach the required conditions, they are used to fine-tune/train the assistant model ([0200], [0182]-[0185]).
Hettige does not disclose wherein the current one of the conversation experiments is updated only upon the successful validation of the corresponding one of the one or more virtual assistant responses,
Kar disclose wherein the current one of the conversation experiments is updated only upon the successful validation of the corresponding one of the one or more virtual assistant responses, (See sections 0032 and 0036-0047: thus The synthetic datasets are validated/improved by comparing against real-world validation sets and using feedback (distribution matching, task validation, see [0032], [0036]-[0047]). Only data that meets validation criteria (distributional similarity, task performance) is retained or used for further training)
Therefore it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to include the teaching of updating the model only when the validation is complete. The motivation is that it make the model to be improved and also handle new inputs.
Claim 22, Hettige in view of Kar (Section 0022- thus grammars reads on rules) discloses wherein the plurality of data inputs comprise: a domain name, a use case name, a use case description, use case attributes, one or more business rules, one or more conversation rules, the one or more exit conditions, details of function calls, one or more sample conversations, a summary of the one or more sample conversations, or one or more conversation templates. (Hettige: Section 0191 Inputs include domain name, use case, business rules, templates, etc. are used as inputs into the virtual assistant- Also see Section 0143, lines 9-12- simple rules-based systems are used as data inputs)
Claim 23, Hettige in view of Kar (Section 0022 and 0023) discloses wherein the plurality of data inputs received from the user device is in natural language text. (Hettige: Section 0143, lines 2 and 6 users through “natural language” and “user input” these are natural text)
Claim 24, Hettige in view of Kar discloses wherein the plurality of data inputs received from the user device is in a structured data format. (Hettige: Section 0152: Templates and data are in structured formats (e.g., JSON)) (Also see 0144)
Claim 25, Hettige in view of Kar (Section 0022-0023) disclose wherein the structured data format is a JavaScript Object Notation (JSON) format. (Hettige: Section 0153, lines 16 “args JSON schema”)
Claim 26, Hettige in view of Kar (Section 0023) discloses wherein the validation data comprises one or more of: the prior version of the current one of the conversation experiments, one or more business rules, one or more conversation rules, the one or more exit conditions, use case attributes, one or more conversation templates, or user sentiment. (Hettige: Sections 0168: Validation uses these prompt and response template See Section 0033, lines 10-15)
Claim 27, Hettige in view of Kar discloses further comprises stored instructions which when executed by the one or more processors prior to the repeating, causes the one or more processors to generate a reason for validation failure when at least one of the one or more subsequent user utterances, the one or more virtual assistant responses, or the one or more virtual assistant actions fails the validation. (Hettige: Section 0131, Validation process includes error analysis and feedback Also see Section 0168]).
Claim 28, Hettige in view of Kar (task performance scores) as feedback for updating the generative model ([0036]-[0047]) discloses wherein the reason for validation failure is used as feedback to generate: one or more subsequent virtual assistant responses, one or more subsequent virtual assistant actions, or the one or more subsequent user utterances, to avoid any further validation failures. (Hettige: Section 0168, the feedback is used to improve synthetic data generation by retraining the neural network)
Claim 29, Hettige in view of Kar discloses wherein the one or more exit conditions comprise: a user utterance or a virtual assistant response comprising one or more keywords indicating completion of a conversation experiment, (Hettige: Section 0198, lines 5-6 …”Generation of synthetic datapoints of a desired size reads on the indication of keywords completion…”) or a validation failure has occurred successively for a threshold number of times for the current one of the conversation experiments. (Hettige: Iterative process stops on exit conditions [0168]).
Claim 30, Hettige in view of Kar discloses wherein the generating and the validating are performed by prompting one or more language models. (Hettige: Sections 0168: Validation uses these prompt and response template See Section 0033, lines 10-15)
Cited Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
He (20160379112) describes systems, methods, and computer-readable media for training computational models such as deep neural networks (DNNs) and for using the trained computational models in, e.g., extrapolating data series. In some examples, a computing device extracts feature values from a plurality of datasets organized according to respective, different data domains, each feature value corresponding to a time.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Akwasi M Sarpong whose telephone number is (571)270-3438. The examiner can normally be reached Mon-Fri. 8:00am-4:00pm.
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.
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.
/AKWASI M SARPONG/SPE, Art Unit 2681
Prosecution Insights