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 .
Allowable Subject Matter
Claims 23, 27-29, 32, 36, and 39 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
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, 21, 22, 24-26, 30, 31, 33-35, 37, and 38 are rejected under 35 U.S.C. 103 as being unpatentable over Thomson et al. (US Pub. 20180330721) in view of Orhan et al. (US Pub. 20240028830).
Referring to claim 1, Thomson discloses A method of operation for a node [fig. 7A, digital assistant system 700] implementing machine learning, ML [pars. 227, 247, and 263; note machine learning (ML)], wherein the node instructs actions in an environment in accordance with a policy generated by a ML agent [figs. 7A and 7B; pars. 208, 246, 253, and 261-263; the digital assistant system performs a policy action selected using one or more policy models (e.g., ML models) of a digital assistant], and wherein the ML agent models the environment [fig. 7B; pars. 208, 246, 253, and 261-263; the digital assistant constructs a belief state comprising current and previous observations of a dialogue and including context data], the method comprising:
obtaining an intent, wherein the intent specifies one or more criteria to be satisfied by the environment [pars. 208, 217-223, 229, 230, and 237; a user intent is inferred from a user request by identifying one or multiple actionable candidate user intents corresponding to the user request and one or more parameters that are populated with specific information and requirements specified in the user request or via contextual information (including context of the user request and context related to the surrounding environment)];
determining an intent cluster from among a plurality of intent clusters to which the intent maps, the determination being based on the criteria specified by the intent [pars. 208, 217-223, 237, 242, 245-247, and 253; the actionable candidate user intents represent candidate domains and/or candidate super-domains (i.e., intent clusters) from an ontology (i.e., a plurality of intent clusters); the relevant candidate user intents are determined by constructing the belief state based on only a subset of nodes in an ontology that are the most relevant; note the parameters including the specific information, the user-specified requirements, and the requirements based on contextual information)];
generating one or more suggested actions to be performed on the environment using the trained ML model [figs. 1 and 7B; pars. 208, 246, 253, and 261-263; one or more candidate policy actions are determined by the policy models (e.g., ML models)].
Thomson does not appear to explicitly disclose setting initialisation parameters for a ML model to be used to model the intent, based on the determined intent cluster; and training the ML model using training data specific to the intent.
However, Orhan discloses setting initialisation parameters for a ML model to be used to model the intent, based on the determined intent cluster [pars. 16 and 17; one or more machine learning models are trained to automate application management tasks; this automation is applicable to a variety of problem domains; instead of training different models for different problem domains, a single model may be trained by including a problem domain (i.e., determined intent cluster) identifier as a hyper-parameter or configuration setting (i.e., initialization parameter)]; and training the ML model using training data specific to the intent [par. 17; for respective problem domains, the machine learning models are trained using relevant domain-specific metadata (i.e., training data specific to the intent)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the digital assistant taught by Thomson, so that the inferred user intent (i.e., the one or multiple actionable intents such as domains or super-domains) are provided as a hyper-parameter or configuration setting to the policy models and the policy models are trained using training data specific to the inferred user intent as taught by Orhan, with a reasonable expectation of success. The motivation for doing so would have been to avoid training different models for different problem domains [Orhan, par. 17].
Referring to claim 21, see at least the rejection for claim 1. Thomson further discloses wherein the node comprises processing circuitry and a memory containing instructions executable by the processing circuitry, whereby the node is operable to perform the claimed steps [fig. 7A; digital assistant system 700 comprises processor(s) 704 and memory 702 for executing digital assistant 726].
Referring to claim 22, Orhan discloses The node of claim 21, configured to obtain the training data specific to the intent using state transition information obtained from the environment [pars. 16-21; the domain-specific metadata includes both current and historical information].
Referring to claim 24, Thomson discloses The node of claim 22, configured to use reinforcement learning, RL, to train the ML model [par. 264; reinforcement learning is used to train the policy models].
Referring to claim 25, Thomson discloses The node of claim 21, configured to determine the intent cluster to which the intent maps by determining the similarity of the one or more criteria of the intent to the criteria of the intents in the plurality of intent clusters [pars. 208, 217-226, 237, 252, and 253; the relevant candidate user intents are determined by mapping the actionable candidate user intents (and vocabulary associated therewith) and the parameters to actionable intent nodes and property nodes, respectively, in the ontology; here, mapping implies matching based on similarity].
Referring to claim 26, Thomson discloses The node of claim 25, wherein the intent is mapped to the intent cluster having the most similar criteria to those of the intent clusters [pars. 208, 217-226, 237, 252, and 253; note the relevant candidate user intents are determined based on only the subset of nodes in an ontology that are the most relevant (i.e., similar)].
Referring to claim 30, Thomson discloses The node of claim 25, configured to determine the intent cluster to which the intent maps by performing an ontological analysis of the intent criteria to determine related criteria to the one or more intent criteria, and utilising the related criteria information to map the intent to an intent cluster [pars. 208, 217-226, 237, 252, and 253; note the mapping of the actionable candidate user intents and the parameters to nodes in the ontology].
Referring to claim 31, Thomson discloses The node of claim 21 further configured to generate the plurality of intent clusters [par. 222; the ontology can be modified by adding domains or nodes or by modifying relationships between nodes within the ontology].
Referring to claim 33, Thomson discloses The node of claim 31, configured to generate the plurality of intent clusters by performing an ontological analysis of the intent criteria of each intent to determine related criteria to the intent criteria, and utilising the related criteria information when generating the plurality of intent clusters [pars. 208, 217-226, 237, 252, and 253; the relevant candidate user intents are determined by mapping the actionable candidate user intents (and vocabulary associated therewith) and the parameters to actionable intent nodes and property nodes, respectively, in the ontology].
Referring to claim 34, Orhan discloses The node of claim 21 configured to determine, for each intent cluster, initialisation parameters [pars. 16 and 17; note the problem domain identifier].
Referring to claim 35, Orhan discloses The node of claim 34, configured to determine the initialisation parameters using multi-task meta learning pre-training [par. 17; the single model is a multi-problem-domain model that is (pre) trained to make predictions for numerous problem domain such that each problem domain is associated with a problem domain identifier].
Referring to claim 37, Thomspon discloses The node of claim 21 further configured to select an action from the one or more suggested actions, and to cause the action to be implemented in the environment [pars. 261-264; a policy action is selected from the candidate policy actions and performed].
Referring to claim 38, Thomson discloses The node of claim 37, wherein the environment is at least a part of a telecommunications network [figs. 1 and 7A; system 100, in which the digital assistant implemented, comprises telephony and other communication services over a network].
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GRACE PARK whose telephone number is (571)270-7727. The examiner can normally be reached M-F 8AM-5PM.
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, TAMARA KYLE can be reached at (571)272-4241. 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.
/Grace Park/Primary Examiner, Art Unit 2144
Prosecution Insights