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 .
Response to Amendment
The Amendment filed 06/05/2026 has been entered. Claims 72-91 remain pending in the application.
Allowable Subject Matter
Claim 77-78, 80-85 and 90-91 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
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.
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 of this title, 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 72-76, 79 and 86-89 are rejected under 35 U.S.C. 103 as being unpatentable over WENCHEL et al. (US 20210174258 A1 hereinafter Wenchel) in view of Yao et al. (US 20210021494 A1) hereinafter Yao and DE BROUWER et al. (US 20200293887 A1) hereinafter De Brouwer.
As to independent claim 72, Wenchel teaches a method performed comprising training a first machine learning inference model using a first training data set; [train models and data used for training ¶17 "combinations of data within the ML system (including one or more of: raw data, processed data, data explicitly used for training ML models"]
detecting unsuitability of the first machine learning inference model for a current network context based on network-related input data associated with the current network context; [monitors model health (unsuitability) for particular dataset ¶17-18 " addressing health and performance issues in live ML-based computer systems. The methods can take into account combinations of data within the ML system (including one or more of: raw data, processed data, data explicitly used for training ML models, data explicitly used for testing ML models, and live data being run through ML models), "]
retraining, upon the detected unsuitability of the first machine learning inference model for the current network context, the first machine learning inference model using a second training data set including at least a portion of the network-related input data; and [resample and retraining triggered by poor health ¶19 "aggregate health score(s) can be used to trigger one or more actions including, but not limited to: generating and sending an alert to one or more users of the ML system, automatically retraining one or more ML models, disabling one or more ML models, taking one or more ML models out of production/operation (e.g., from one or more processors), re-sampling training data to address one or more identified problem areas). "]
transmitting, to the artificial intelligence or machine learning model consumer, a second machine learning model message including the re-trained first machine learning inference model. [trigger redeployment ¶78 " trigger an automated/automatic redeployment process 722 that causes the user(s)' newly-retrained model to the model server 704."]
Wenchel does not specifically teach a core network function for a communication system with a current network context;
However, Yao teaches a core network function for a communication system with a current network context; [network functions (nwdaf g5c) ¶46-47 and training models ¶51]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the model training disclosed by Wenchel by incorporating the a core network function for a communication system with a current network context disclosed by Yao because both techniques address the same field of machine learning and by incorporating Yao into Wenchel improve network performance and efficiency in machine learning [Yao ¶3, ¶31].
Wenchel and Yao do not specifically teach transmitting, to an artificial intelligence or machine learning model consumer, a first machine learning model message including the first machine learning inference model;
However, De Brouwer teaches transmitting, to an artificial intelligence or machine learning model consumer, a first machine learning model message including the first machine learning inference model; [messages between coordinating server and edge devices (consumer) with first models ¶12 "a coordinating server 1221 that manages training tasks and performs model aggregation. Below the coordinating server, the figure illustrates four edge devices 953 that perform training using local data to produce local updates of a base model. In the figure, messages travel down and up between the coordinating server 1221 and individual devices 953, each represented by a horizontal line."]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the model training disclosed by Wenchel and Yao by incorporating the transmitting, to an artificial intelligence or machine learning model consumer, a first machine learning model message including the first machine learning inference model disclosed by De Brouwer because both techniques address the same field of machine learning and by incorporating De Brouwer into Wenchel and Yao improve training of models with privacy, security and ownership of data [De Brouwer ¶13]
.
As to dependent claim 73, the rejection of claim 72 is incorporated, Wenchel, Yao and De Brouwer further teach transmitting, to a data producer, a first data collection request; [De Brouwer messages between coordinating server and edge devices (consumer) with first models ¶12 ]
receiving, from the data producer, a first data collection response including first collected data, and generating the first training data set based on the first collected data. [Wenchel sampling is a request for training data for a model ¶20 " data used to train ML models (“training data”) may be sampled, possibly via a noisy process, from some underlying and potentially opaque true data distribution or source."]
As to dependent claim 74, the rejection of claim 72 is incorporated, Wenchel, Yao and De Brouwer further teach further comprising: training a first machine learning monitoring model based on the first training data set. [Wenchel training a model ¶76 "training pipeline 705 is a process that selects, trains, creates, or generates AI models."], [Wenchel model is for monitoring and outputs explanations ¶44 " model outputs 114 can include explanations of the ML model inferences. The explanations c
As to dependent claim 75, the rejection of claim 74 is incorporated, Wenchel, Yao and De Brouwer further teach further comprising: transmitting the network related input data to the artificial intelligence or machine learning model consumer, and [Wenchel could network receives input data ¶77 "an API in the multi-stage system cloud network 702 that receives new feature values (e.g. including input data, such as input data 608"]
feeding the first machine learning monitoring model with the network related input data. [Wenchel updates model accordingly ¶77 "The model endpoint 710 can update one or more of at least the following components of FIG. 6: explainer 610, model 601A, reference data 606"]
As to dependent claim 76, the rejection of claim 75 is incorporated, Wenchel, Yao and De Brouwer further teach wherein the network-related input data comprises second collected data received in a second data collection response in response to a second data collection request. [Yao event data collection ¶43-44]
As to dependent claim 79, the rejection of claim 74 is incorporated, Wenchel, Yao and De Brouwer further teach wherein the first machine learning model message further includes the first machine learning monitoring model. [Wenchel sends models to the cloud ¶76 "a user model, model metadata associated with the user model, and one or more explainers are sent to the multi-stage system cloud network 702."]
As to dependent claim 86, the rejection of claim 72 is incorporated, Wenchel, Yao and De Brouwer further teach an apparatus comprising at least one processor, at least one memory including computer program code, and at least one interface configured for communication with at least another apparatus, the at least one processor, with the at least one memory and the computer program code, being configured to cause the apparatus to perform the method of claim 72. [Wenchel processor, interfaces, memory and instructions ¶81-82]
As to independent claim 87, Wenchel teaches a method performed by a user equipment, the method comprising [user cloud network ¶76]
receiving a first machine learning inference model, [receives models in the cloud ¶76 "a user model, model metadata associated with the user model, and one or more explainers are sent to the multi-stage system cloud network 702."] , [inference ¶59]
obtaining network related input data, [generates data from a stream ¶4, ¶77 "receives new feature values (e.g. including input data, such as input data 608 in FIG. 6) from the multi-stage system interface 706" ]
feeding the first machine learning inference model with the network related input data; [training data for model (feeds model) ¶20 " data used to train ML models (“training data”) may be sampled, possibly via a noisy process, from some underlying and potentially opaque true data distribution or source."]
transmitting, upon detected unsuitability of said first machine learning inference model for said current network context based on said network-related input data, a machine learning model retraining request message; [monitors model health (unsuitability) for particular dataset ¶17-18 " addressing health and performance issues in live ML-based computer systems. The methods can take into account combinations of data within the ML system (including one or more of: raw data, processed data, data explicitly used for training ML models, data explicitly used for testing ML models, and live data being run through ML models), "]
receiving, from the core network function, a second machine learning model message including a re-trained version of the first machine learning inference model, and [resample and retraining triggered by poor health ¶19 "aggregate health score(s) can be used to trigger one or more actions including, but not limited to: generating and sending an alert to one or more users of the ML system, automatically retraining one or more ML models, disabling one or more ML models, taking one or more ML models out of production/operation (e.g., from one or more processors), re-sampling training data to address one or more identified problem areas). "]
replacing the first machine learning inference model with the re-trained version of the first machine learning inference model. [trigger redeployment ¶78 " trigger an automated/automatic redeployment process 722 that causes the user(s)' newly-retrained model to the model server 704."]
Wenchel does not specifically teach a core network function for a communication system with a current network context;
However, Yao teaches a core network function for a communication system with a current network context; [network functions (nwdaf g5c) ¶46-47 and training models ¶51], [network status/events (context) ¶30]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the model training disclosed by Wenchel by incorporating a core network function for a communication system with a current network context disclosed by Yao because both techniques address the same field of machine learning and by incorporating Yao into Wenchel improve network performance and efficiency in machine learning [Yao ¶3, ¶31].
Wenchel and Yao do not specifically teach a first machine learning model message including a first machine learning inference model;
However, De Brouwer teaches a first machine learning model message including a first machine learning inference model; [messages between coordinating server and edge devices (consumer) with first models ¶12 "a coordinating server 1221 that manages training tasks and performs model aggregation. Below the coordinating server, the figure illustrates four edge devices 953 that perform training using local data to produce local updates of a base model. In the figure, messages travel down and up between the coordinating server 1221 and individual devices 953, each represented by a horizontal line."]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the model training disclosed by Wenchel and Yao by incorporating the a first machine learning model message including a first machine learning inference model disclosed by De Brouwer because both techniques address the same field of machine learning and by incorporating De Brouwer into Wenchel and Yao improve training of models with privacy, security and ownership of data [De Brouwer ¶13]
As to dependent claim 88, the rejection of claim 87 is incorporated, Wenchel, Yao and De Brouwer further teach in relation to said obtaining, the method further comprises receiving said network related input data. [Wenchel generates data from a stream ¶4, ¶77 "receives new feature values (e.g. including input data, such as input data 608 in FIG. 6) from the multi-stage system interface 706"]
As to dependent claim 89, the rejection of claim 87 is incorporated, Wenchel, Yao and De Brouwer further teach wherein a first machine learning monitoring model is included by [[a]] the first machine learning model message, and wherein the obtaining comprises collecting said network related input data. [Wenchel generates data from a stream ¶4, ¶77 "receives new feature values (e.g. including input data, such as input data 608 in FIG. 6) from the multi-stage system interface 706" ]
Response to Arguments
Applicant's arguments filed 06/05/2026, with respect to 101 these rejection have been withdrawn.
Applicant's arguments filed 06/05/2026. In the remark, applicant argues that:
Wenchel fails to teach the core network function, as recited in independent Claim 72, identifies the entity that performs the claimed training, transmitting, detecting, retraining, and transmitting operations. Wenchel's generic cloud-based ML monitoring system is not a core network function. Wenchel does not disclose a 5G core network function, a 5GS entity, an NWDAF, an MDAF, an AMF, or any other core network function that trains and distributes machine learning inference models to an AI/ML model consumer. At most, Wenchel discloses generic computer-network or cloud infrastructure for monitoring ML systems. Such generic infrastructure is not the claimed core network function as recited in amended claim 72.
Wenchel also does not disclose transmitting, by a core network function, a first machine learning model message to an artificial intelligence or machine learning model consumer as provided in independent Claim 72. Rather, Wenchel describes registering or onboarding user models, model metadata, and explainers with a monitoring platform, which is materially different from a core network function transmitting a first machine learning model message including a first machine learning inference model to an AI/ML model consumer in a communication system as provided in Claim 72. Wenchel's model server, training pipeline, and monitoring platform do not correspond to the claimed model-provider/core-network- function architecture.
Wenchel further fails to disclose detecting unsuitability of the first machine learning inference model for a current network context based on network-related input data associated with the current network context as provided in independent Claim 72. The claimed "current network context" and "network-related input data" are not satisfied by Wenchel's generic ML data streams, health metrics, data drift metrics, accuracy metrics, fairness metrics, bias metrics, explainability metrics, or dashboard alerts. That is, while Wenchel may disclose detecting generic ML performance issues such as data drift, overfitting, bias, fairness, or degraded accuracy, Wenchel does not disclose detecting that a machine learning inference model is-8-
unsuitable for a current network context of a communication system based on network-related input data associated with that current network context.
Still further, Wenchel fails to disclose retraining the first machine learning inference model upon the claimed detected unsuitability for the current network context using a second training data set including at least a portion of the network-related input data as provided in independent Claim 72. Wenchel's generic disclosure of automatic retraining or redeployment in response to generic ML monitoring alerts does not anticipate the specifically claimed trigger and training data. Claim 72 provides for retraining to occur upon detecting unsuitability for the current network context, and the second training data set to include at least a portion of the network-related input data associated with that current network context. Wenchel does not disclose that relationship.
Finally, Wenchel does not disclose transmitting, by the core network function to the AI/ML model consumer, a second machine learning model message including the re-trained first machine learning inference model as provided in independent Claim 72. Wenchel's disclosure that a monitoring platform may trigger automatic retraining or redeployment of a model to a model server does not disclose the claimed second machine learning model message from a core network function to an AI/ML model consumer.
As to points (1)-(6), Applicant’s arguments with respect to claim 72 have been considered but are moot in view of a new ground of rejection as set forth above of Wenchel in view of Yao and De Brouwer.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 C.F.R. § 1.111(c) to consider these references fully when responding to this action.
GHANTA et al. (US 20190377984 A1) teaches detecting suitability of models and data networks (See ¶34).
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BEAU SPRATT whose telephone number is (571)272-9919. The examiner can normally be reached M-F 8:30-5 PST.
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, Jennifer Welch can be reached on 5712127212. 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.
/BEAU D SPRATT/Primary Examiner, Art Unit 2143
Prosecution Insights