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 .
Claim Rejections - 35 USC § 102
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 (i.e., changing from AIA to pre-AIA ) 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.
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.
Claims 1-30 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Li et al. (US 2025/0219898).
Regarding claims 1 and 30, Li discloses a user equipment (UE) 200 in FIG. 3 and a method for wireless communication by the UE, comprising: one or more memories storing processor-executable code; and one or more processors coupled with the one or more memories and individually or collectively operable to execute the code to cause the UE to:
receive one or more control messages that indicate a life cycle management operation for a machine learning model or machine learning model-based functionality associated with the UE (paragraph 140, lines 2-3: “the network (e.g., a network node of the network) configures the UE with an operation of the ML model (at operation 404)”), wherein the one or more control messages comprise an indication of whether the life cycle management operation is based at least in part on a performance of the machine learning model or machine learning model-based functionality (paragraph 140, line 3: “the network configures the UE with an ML model performance indication (at operation 406)”); and
perform the life cycle management operation for the machine learning model or machine learning model-based functionality based at least in part on the one or more control messages (paragraph 140, lines 4-5: “the UE performs the requested ML model performance analysis (at operation 410)”).
Regarding claim 2, Li discloses that the indication further comprises a cause of the life cycle management operation, the cause comprising at least one of the performance of the machine learning model or machine learning model-based functionality, a network associated with the UE being overloaded, and a network associated with the UE transitioning to an energy-saving mode (paragraph 140, lines 6-7: “the network indicates to the UE to begin using a non-ML based algorithm (at operation 414) when the network determines performance is likely to be improved by the change.”).
Regarding claim 3, Li discloses that the indication further comprises a performance report for the machine learning model or machine learning model-based functionality based at least in part on the life cycle management operation (paragraph 196, lines 1-2: “The NW may configure the UE with configuration(s) related to the event/condition-based aperiodic ML-model performance report (e.g., upon reception of an RRC message)”).
Regarding claim 4, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the UE to: transmit, to a network entity, a message including a request for the performance report, wherein receiving the one or more control messages is based at least in part on the request (paragraph 177, lines 1-2 and paragraph 178: “The NW trigger/request message (e.g., MAC CE, RRC or L1 signaling) preferably includes at least one indication or reference of one or more of the following: Which ML model(s) and/or ML features to report the performance for”).
Regarding claim 5, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the UE to: transmit, within the message, an indication of one or more parameters for the performance report, the one or more parameters comprising one or more performance metrics associated with the machine learning model or machine learning model-based functionality that are to be included in the performance report, a time duration for the network entity to monitor the one or more performance metrics, or a combination thereof (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 6, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the UE to: transmit the performance report to a server associated with the machine learning model or machine learning model-based functionality based at least in part on receiving the one or more control messages (paragraph 271, lines 6-8: “Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS)”).
Regarding claim 7, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the UE to: receive the performance report from a network entity, wherein transmitting the performance report is based at least in part on receiving the performance report (paragraph 48, lines 4-5: “At an operation 274, the network node 210 may send (and the UE 200 may receive) an indication to stop using a currently used ML model.”).
Regarding claim 8, Li discloses that the performance report comprises at least one of an indication of one or more performance metrics associated with the machine learning model or machine learning model-based functionality, an indication of a time duration during which the one or more performance metrics were monitored, an indication of a network operation that triggered the life cycle management operation, an indication of a network performance metric that triggered the life cycle management operation, or a combination thereof (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 9, Li discloses that the performance report comprises one or more statistics corresponding to the performance of the machine learning model or machine learning model-based functionality for a time duration, the one or more statistics including an indication of an occurrence rate of life cycle management operations for the machine learning model or machine learning model-based functionality, a mapping between the performance of the machine learning model or machine learning model-based functionality and a set of life cycle management operations including the life cycle management operation, a list of one or more machine learning models or machine learning model-based functionalities to which the UE switched in accordance with the life cycle management operation, an indication of a respective occurrence rate associated with the UE switching to each machine learning model or machine learning model-based functionality of the one or more machine learning models or machine learning model-based functionalities, or a combination thereof (paragraph 146, lines 3-4: “the input data distributions as well as the mapping between the input data and output data of the ML-model holds for a relative long period, e.g., in a few days.”).
Regarding claim 10, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the UE to: generate a performance report for the machine learning model or machine learning model-based functionality based at least in part on the indication that the life cycle management operation is based at least in part on the performance of the machine learning model or machine learning model-based functionality, the performance report comprising an identifier of the machine learning model or machine learning model-based functionality based at least in part on the machine learning model or machine learning model-based functionality being active when the one or more control messages are received, an indication of a functionality of the machine learning model based at least in part on the functionality being active when the one or more control messages are received, an occurrence rate of life cycle management operations for the machine learning model or machine learning model-based functionality, a list of one or more machine learning models or machine learning model-based functionalities to which the UE switched in accordance with the life cycle management operation, or a combination thereof (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 11, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the UE to: transmit the performance report to a server associated with the machine learning model or machine learning model-based functionality based at least in part on receiving the one or more control messages (paragraph 271, lines 6-8: “Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS)”).
Regarding claim 12, Li discloses that the indication comprises a single bit, a first value of the bit corresponds to an indication that the life cycle management operation is based at least in part on the performance of the machine learning model or machine learning model-based functionality, and a second value of the bit corresponds to an indication that the life cycle management operation is based at least in part on a performance metric of a network to which the UE belongs (paragraph 184, lines 8-11: “the network node may indicate the UE to perform aperiodic reporting based on, e.g., a value in percentage that indicates the confidence level of its ML-model output, a single bit indicating trust or no trust in the model, a single bit indicating that the input data is out-of-distribution, or a metric specific to the current model that ML performance is monitored for.”).
Regarding claim 13, Li discloses that the indication further comprises information about a decision, by a network entity, to indicate the life cycle management operation to the UE (paragraph 260, lines 1-2: “Based on the ML-model performance analysis, the NW decides and signals to the UE, e.g., to stop using the ML model”).
Regarding claim 14, Li discloses a network entity, comprising: one or more memories storing processor-executable code; and one or more processors coupled with the one or more memories and individually or collectively operable to execute the code to cause the network entity to:
select a life cycle management operation to be performed by a user equipment (UE) for a machine learning model or machine learning model-based functionality at the UE (paragraph 141, line 2: “the network configures the UE with or for operation of a selected ML model”); and
transmit, to the UE, one or more control messages that indicate the life cycle management operation for the machine learning model or machine learning model-based functionality (paragraph 140, lines 2-3: “the network (e.g., a network node of the network) configures the UE with an operation of the ML model (at operation 404)”), wherein the one or more control messages comprise an indication of whether the life cycle management operation is based at least in part on a performance of the machine learning model or machine learning model-based functionality.
Regarding claim 15, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the network entity to: monitor one or more performance metrics associated with the performance of the machine learning model or machine learning model-based functionality, wherein the life cycle management operation is selected based at least in part on the one or more performance metrics, and wherein the one or more control messages comprise the indication that the life cycle management operation is based at least in part on the performance of the machine learning model or machine learning model-based functionality (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 16, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the network entity to: monitor one or more network performance metrics associated with a network to which the network entity and the UE belong, wherein the life cycle management operation is selected based at least in part on the one or more network performance metrics, and wherein the one or more control messages comprise the indication that the life cycle management operation is based at least in part on the one or more network performance metrics (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 17, Li discloses that the life cycle management operation is selected in accordance with a decision, by a network associated with the network entity and the UE, to operate in an energy-saving mode, and the one or more control messages comprise the indication that the life cycle management operation is based at least in part on the decision (paragraph 322: “the teachings of these embodiments may improve … power consumption, and thereby provide benefits such as … extended battery lifetime.”).
Regarding claim 18, Li discloses that, to transmit the one or more control messages, the one or more processors are individually or collectively operable to execute the code to cause the network entity to: generate a performance report for the machine learning model or machine learning model-based functionality based at least in part on the life cycle management operation; and transmit the indication comprising the performance report (paragraph 140, line 3: “the network configures the UE with an ML model performance indication (at operation 406)”).
Regarding claim 19, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the network entity to: receive a message including a request for the performance report, wherein transmitting the one or more control messages is based at least in part on the request (paragraph 177, lines 1-2 and paragraph 178: “The NW trigger/request message (e.g., MAC CE, RRC or L1 signaling) preferably includes at least one indication or reference of one or more of the following: Which ML model(s) and/or ML features to report the performance for”).
Regarding claim 20, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the network entity to: receive a message indicating one or more parameters for the performance report, the one or more parameters comprising one or more performance metrics associated with the machine learning model or machine learning model-based functionality that are to be included in the performance report, a time duration for the network entity to monitor the one or more performance metrics, or a combination thereof, wherein the indication comprising the performance report is transmitted in accordance with the one or more parameters (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 21, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the network entity to: transmit the performance report to a server associated with the machine learning model or machine learning model-based functionality based at least in part on transmitting the one or more control messages (paragraph 271, lines 6-8: “Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS)”).
Regarding claim 22, Li discloses that the performance report comprises at least one of an indication of one or more performance metrics associated with the machine learning model or machine learning model-based functionality, an indication of a time duration during which the one or more performance metrics were monitored, an indication of a network operation that triggered the life cycle management operation, an indication of a network performance metric that triggered the life cycle management operation, or a combination thereof (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 23, Li discloses that the performance report comprises one or more statistics corresponding to the performance of the machine learning model or machine learning model-based functionality for a time duration, the one or more statistics including an indication of an occurrence rate of life cycle management operations for the machine learning model or machine learning model-based functionality, a mapping between the performance of the machine learning model or machine learning model-based functionality and a set of life cycle management operations including the life cycle management operation, a list of one or more machine learning models or machine learning model-based functionalities to which the UE switched in accordance with the life cycle management operation, an indication of a respective occurrence rate associated with the UE switching to each machine learning model or machine learning model-based functionality of the one or more machine learning models or machine learning model-based functionalities, or a combination thereof (paragraph 146, lines 3-4: “the input data distributions as well as the mapping between the input data and output data of the ML-model holds for a relative long period, e.g., in a few days.”).
Regarding claim 24, Li discloses that the indication comprises a single bit, a first value of the bit corresponds to an indication that the life cycle management operation is based at least in part on the performance of the machine learning model or machine learning model-based functionality, and a second value of the bit corresponds to an indication that the life cycle management operation is based at least in part on a performance metric of a network to which the UE belongs (paragraph 184, lines 8-11: “the network node may indicate the UE to perform aperiodic reporting based on, e.g., a value in percentage that indicates the confidence level of its ML-model output, a single bit indicating trust or no trust in the model, a single bit indicating that the input data is out-of-distribution, or a metric specific to the current model that ML performance is monitored for.”).
Regarding claim 25, Li discloses a server (paragraph 271, lines 6-8: “Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS)”), comprising: one or more memories storing processor-executable code; and one or more processors coupled with the one or more memories and individually or collectively operable to execute the code to cause the server to:
transmit, to a user equipment (UE), a machine learning model configuration for a machine learning model or machine learning model-based functionality of the UE (paragraph 140, line 3: “the network configures the UE with an ML model performance indication (at operation 406)”);
receive a performance report associated with the machine learning model or machine learning model-based functionality based at least in part on a life cycle management operation for the machine learning model or machine learning model-based functionality, the performance report indicating one or more performance metrics associated with a performance of the machine learning model or machine learning model-based functionality (paragraph 140, lines 4-6: “the UE performs the requested ML model performance analysis (at operation 410); the UE indicates (transmits and indicator or indication) of the ML model performance to the network (at operation 412)”); and
update the machine learning model or machine learning model-based functionality based at least in part on the performance report (paragraph 259, line 1: “the network node stores the received periodical ML-model performance report”).
Regarding claim 26, Li discloses that the one or more processors are individually or collectively further operable to execute the code to cause the server to: transmit, to a network entity, a message including a request for the performance report and including an indication of one or more parameters for the performance report, the one or more parameters comprising one or more performance metrics associated with the machine learning model or machine learning model-based functionality that are to be included in the performance report, a time duration for the network entity to monitor the one or more performance metrics, or a combination thereof, wherein the performance report is received from the network entity based at least in part on the request (paragraph 177, lines 1-2 and paragraph 178: “The NW trigger/request message (e.g., MAC CE, RRC or L1 signaling) preferably includes at least one indication or reference of one or more of the following: Which ML model(s) and/or ML features to report the performance for”).
Regarding claim 27, Li discloses that the performance report comprises at least one of an indication of one or more performance metrics associated with the machine learning model or machine learning model-based functionality, an indication of a time duration during which the one or more performance metrics were monitored, an indication of a network operation that triggered the life cycle management operation, an indication of a network performance metric that triggered the life cycle management operation, or a combination thereof (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Regarding claim 28, Li discloses that the performance report comprises one or more statistics corresponding to the performance of the machine learning model or machine learning model-based functionality for a time duration, the one or more statistics including an indication of an occurrence rate of life cycle management operations for the machine learning model or machine learning model-based functionality, a mapping between the performance of the machine learning model and a set of life cycle management operations including the life cycle management operation, a list of one or more machine learning models or machine learning model-based functionalities to which the UE switched in accordance with the life cycle management operation, an indication of a respective occurrence rate associated with the UE switching to each machine learning model or machine learning model-based functionality of the one or more machine learning models or machine learning model-based functionalities, or a combination thereof (paragraph 146, lines 3-4: “the input data distributions as well as the mapping between the input data and output data of the ML-model holds for a relative long period, e.g., in a few days.”).
Regarding claim 29, Li discloses that, to receive the performance report, the one or more processors are individually or collectively operable to execute the code to cause the server to: receive the performance report from the UE, the performance report comprising an identifier of the machine learning model or machine learning model-based functionality, an indication of a functionality of the machine learning model or machine learning model-based functionality, an occurrence rate of life cycle management operations for the machine learning model or machine learning model-based functionality, a list of one or more machine learning models or machine learning model-based functionalities to which the UE switched in accordance with the life cycle management operation, or a combination thereof (paragraph 161, lines 1-3: “The reporting quantity for the ML-model performance monitoring in case there may be multiple performance metrics the UE is capable of measuring (there may be different UE capabilities reported to the network, based on what the UE is capable of measuring in terms of metric for ML-model performance monitoring)”).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Zhu et al. (US 2023/0100253) discloses that a CU-XP transmits a model setup request to a network entity, which runs a model. The model setup request may include a model ID and a parameter set ID.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SAM BHATTACHARYA whose telephone number is (571)272-7917. The examiner can normally be reached weekdays, 9-5:30.
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, Matthew D. Anderson can be reached at (571) 272-4177. 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.
/SAM BHATTACHARYA/Primary Examiner, Art Unit 2646