DETAILED ACTION
This office action is in response to the application filed on 7/26/2024 in which claims 1-16 are pending.
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
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.
Claim(s) 1-16 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Vandikas et al. (US20250047570A1).
As to claims 1 and 11, Vandikas teaches an apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: ([0285] FIG. 13 depicts two different examples in panels a) and b), respectively, of the arrangement that the third node 113 may comprise to perform the method actions described above in relation to FIG. 4, and/or FIGS. 7-8.)
obtain, from a network entity, at least one machine learning model for training or retraining based on measurement data of a network; ([0164] In this Action 401, the third node 113 may obtain, from the fourth node 114 operating in the communications system 100, the second predictive ML model of the expected respective representation of the distribution of data sets corresponding to the one or more first features. [0166] The obtaining in this Action 401 of the second predictive ML model may be in response to the request that the third node 113 may have sent. [0231] fig. 7 and [0232] FIG. 8 is another signalling diagram depicting another non-limiting example of a method performed by the first node 111, the second node 112, the third node 113 and the fourth node 114 according to embodiments herein. Particularly, FIG. 8, spreading over panel a) and panel b), depicts a non-limiting implementation of the methods according to Phase 2. In a same manner as phase 1, the process begins at 1, when a blueprint for the training of the first predictive ML model is obtained by, e.g., submitted to, the third node 113 [0287]- [0289] According to a third option, the second indication may be configured to comprise the metric configured to indicate the change in the respective representation of the distribution. [0290] The third node 113 is also configured to, e.g. by means of a retraining unit 1302 within the third node 113 configured to, retrain, using ML, the first predictive ML model based on the second indication configured to be obtained.)
determine that data collection is required prior to training or retraining the at least one machine learning model, and ([0231] At 2, the third node 113 may then try to collect the data that may be needed to train that ML model using the expected features and the expected output that may have been declared in the ML model blueprint. At 3, the third node 113 may have a local copy of that data, represented as “data exists”. In this case it may not reach out to the first nodes 111, the, eNBs, for the needed data. At steps 4-11, if the third node 113 does not have a local copy of that data it may then reach out to the data sources UE_1 and UE_2, and use the get_data ( ) call to retrieve this information from eNB1 to eNB2 at Step 4 and Step 8, respectively.)
receive, from a network function, one or more measurement reports that comprises at least one dataset from the data collection; ( [0231] At Step 5 and Step 9, each of eNB1 and eNB2 may respectively instruct UE_1 and UE_2, respectively, to get the measurements. At Step 6 and Step 10, each node may respectively provide back a response, e.g., comprised of one or more IP packets, which are not annotated since in this phase such input is absent in the Data Quality Registry node. At Step 12, the third node 113 has now enough data to train the first predictive ML model which has been declared in the ML model blueprint thus fulfilling the initial purpose of this task [0232] Accordingly, at Step 10, the data and the flag are now transmitted to the third node 113 according to Action 207 and Action 404. At Steps 11-15, the same Steps described in Step 4-10 are repeated for another first node 111, eNB2. That is, the annotated data collection process continues until Step 15. For simplicity, for UE_2, it is considered that no issues were detected by the second predictive ML model. Hence, the data is transmitted to the third node 113 in Step 15, according to Action 207 and Action 404.)
determine whether additional assisted information from one or more network devices is required for training or retraining; and ([0231] At 2, the third node 113 may then try to collect the data that may be needed to train that ML model using the expected features and the expected output that may have been declared in the ML model blueprint. At 3, the third node 113 may have a local copy of that data, represented as “data exists”. In this case it may not reach out to the first nodes 111, the, eNBs, for the needed data. At steps 4-11, if the third node 113 does not have a local copy of that data it may then reach out to the data sources UE_1 and UE_2, and use the get_data ( ) call to retrieve this information from eNB1 to eNB2 at Step 4 and Step 8, respectively At Step 12, the third node 113 has now enough data to train the first predictive ML model which has been declared in the ML model blueprint thus fulfilling the initial purpose of this task.)
train or retrain the at least one machine learning model based on all collected datasets, which comprises the at least one dataset from the data collection. ([0231] At Step 12, the third node 113 has now enough data to train the first predictive ML model which has been declared in the ML model blueprint thus fulfilling the initial purpose of this task. [0232] Step 16, is now conditioned based on the flag. If the data distribution has changed, as was detected early in the data transfer, the third node 113 may need to now retrain the first predictive ML model, according to Action 405, and recompute the input that may be needed for the fourth node 114.)
As to claims 2 and 12, Vandikas teaches the apparatus according to claim 1, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: receive one or more additional models to be trained or retrained in the future, wherein the obtained machine learning model for training or retraining is selected from a plurality of machine learning models accessible to the apparatus. ([0184] this may be understood to mean that the third node 113 may retrain with one or more further iterations the first predictive ML model and recompute the input that may be needed for the fourth node 114, [0232] At Step 3, the second predictive ML model for the specific ML model blueprint may be requested from the fourth node 114. Step 16, is now conditioned based on the flag. If the data distribution has changed, as was detected early in the data transfer, the third node 113 may need to now retrain the first predictive ML model, according to Action 405, and recompute the input that may be needed for the fourth node 114. Step 20 may be understood to be the same as Step 16 in phase 1. [0231] At Step 16, using this input, the data distribution, which may be understood to be a vector, may be used to compare the similarity between the expected distribution and what may be transferred over the communications system 100 to train, according to Action 501, the second predictive ML model to detect any anomalies in the data that may be being transmitted. In another example Step 1 may be omitted and instead of capturing the initial expectation about each data distribution from a live network, the data distribution that was used when training the original ML model in the laboratory may be used in order to use that as a baseline and thus allow for ascertaining that expectation through the lifecycle of the first predictive ML model and to determine when the first predictive ML model may need to be retrained.)
As to claims 3 and 13, Vandikas teaches the apparatus according to claim 1, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: transmit an acknowledgement to the network entity from which the apparatus obtained the at least one machine learning model for training or retraining. ([0163] In some embodiments, the third node 113 may request a specific ML model blueprint of the second predictive ML model, that is, the data quality ML model, from the fourth node 114. [0164] In this Action 401, the third node 113 may obtain, from the fourth node 114 operating in the communications system 100, the second predictive ML model [0185] In this Action 406, the third node 113 may send, to the fourth node 114, the sixth indication.)
As to claims 4 and 14, Vandikas teaches the apparatus according to claim 1, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: validate the trained model based on the one or more measurement reports. ([0026] For example, embodiments herein may be understood to enable prioritization of measurement reports, which may be understood to be responsible for building dataset for training/updating ML models, [0087] Action 201 may be understood to be an optional action. In some examples, instead of capturing the initial expectation about each data distribution from a live network, the data distribution that may have been used when training the original ML model of the first predictive ML model in a laboratory may be used. In order to use that original ML model as a baseline for ascertaining that expectation through the model's lifecycle and to determine when the baseline ML model may need to be retrained. The original ML model may be understood to refer to the very first version of the ML model that may have had high enough performance to be released. As such, it may be considered as a baseline.)
As to claims 5 and 15, Vandikas teaches the apparatus according to claim 1, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: provide the trained model to a network device or a user equipment. ([0231] At Step 15, the source/destination and corresponding ports, the ML model blueprint, the top features, produced from the feature importance function, and the data distribution of the top features may be assembled in a tuple <source, destination, source port, destination port, mb, top_features, data_distribution> and may be recorded in the fourth node 114. [0232] Step 19 may be understood to be the same as Step 15 in phase 1. [0080] According to the foregoing, in this Action 201, the first node 111 may obtain, from the third node 113, e.g., an OAM node, the other ML predictive model, which is referred to herein as a second predictive ML model. 0084] The second predictive ML model may have been determined by the fourth node 114, and may be based on supervised learning, e.g., via a Deep Neural Network (DNN) or reinforcement learning, e.g., via Double Deep-Q Network (DDQN). The second predictive ML model will be described in detail later, in relation to FIG. 5.)
As to claims 6 and 16, Vandikas teaches the apparatus according to claim 1, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: receive, from the network entity, a request for retraining the at least one machine learning model; (fig. 8B if flag retrain model the flag comes from node 111 In another example Step 1 may be omitted and instead of capturing the initial expectation about each data distribution from a live network, the data distribution that was used when training the original ML model in the laboratory may be used in order to use that as a baseline and thus allow for ascertaining that expectation through the lifecycle of the first predictive ML model and to determine when the first predictive ML model may need to be retrained.)
transmit, to the network function, a request for training data for the at least one machine learning model to be retrained; and ([0184] In this Action 405, the third node 113 retrains, using ML, the first predictive ML model based on the obtained second indication. This may be understood to mean that the third node 113 may retrain with one or more further iterations the first predictive ML model and recompute the input that may be needed for the fourth node 114, using the one or more first sets of data, with the proviso that the one or more first sets of data have been received, which may be understood to mean that the one or more first sets of data were of sufficient quality to be used to train the first predictive ML model. otherwise this may mean that the third node 113 may refrain from training the first predictive ML model, with the proviso that the flag of the metric is received instead, indicating that the data quality is too poor to be transmitted to the third node 113 and used to train the first predictive ML model. Fig. 8 step 2-15)
retrain the at least one machine learning model based on the requested training data. (Fig. 8 step 16)
As to claim 7, Vandikas teaches an apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: ([0285] FIG. 13 depicts two different examples in panels a) and b), respectively, of the arrangement that the third node 113 may comprise to perform the method actions described above in relation to FIG. 4, and/or FIGS. 7-8.)
determine that at least one machine learning model requires updates; ([0185] If there is no problem on the side of the second node 112 that may be reported and the flagged dataset may be safely re-used for retraining the first predictive ML model, an indication that the data distribution has changed may need to be updated in the fourth node 114.)
transmit, to a network entity, a request for training or retraining the at least one machine learning model; ([0185] In this Action 406, the third node 113 may send, to the fourth node 114, the sixth indication The sixth indication may further indicate the second predictive ML model.)
provide, to the network entity, at least one measurement report; and ([0186] For example, the sixth indication may be a tuple wherein the source/destination and corresponding ports, the ML model blueprint, the top features, produced from the feature importance function, and the data distribution of the top features may be assembled e.g., as <source, destination, source port, destination port, mb, top_features, data_distribution>.)
receive, from the network entity, a trained or retrained machine learning model to be executed on the apparatus. ([0164] In this Action 401, the third node 113 may obtain, from the fourth node 114 operating in the communications system 100, the second predictive ML model)
As to claim 8, Vandikas teaches the apparatus according to claim 7, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: receive one or more additional machine learning models to be trained or retrained in the future, wherein the obtained machine learning model for training or retraining is selected from a plurality of machine learning models accessible to the apparatus. ([0184] this may be understood to mean that the third node 113 may retrain with one or more further iterations the first predictive ML model and recompute the input that may be needed for the fourth node 114, [0232] At Step 3, the second predictive ML model for the specific ML model blueprint may be requested from the fourth node 114. Step 16, is now conditioned based on the flag. If the data distribution has changed, as was detected early in the data transfer, the third node 113 may need to now retrain the first predictive ML model, according to Action 405, and recompute the input that may be needed for the fourth node 114. Step 20 may be understood to be the same as Step 16 in phase 1. [0231] At Step 16, using this input, the data distribution, which may be understood to be a vector, may be used to compare the similarity between the expected distribution and what may be transferred over the communications system 100 to train, according to Action 501, the second predictive ML model to detect any anomalies in the data that may be being transmitted. In another example Step 1 may be omitted and instead of capturing the initial expectation about each data distribution from a live network, the data distribution that was used when training the original ML model in the laboratory may be used in order to use that as a baseline and thus allow for ascertaining that expectation through the lifecycle of the first predictive ML model and to determine when the first predictive ML model may need to be retrained.)
As to claim 9, Vandikas teaches the apparatus according to claim 7, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: validate the trained machine learning model based on the one or more measurement reports; and execute the trained machine learning model. ([0026] For example, embodiments herein may be understood to enable prioritization of measurement reports, which may be understood to be responsible for building dataset for training/updating ML models, [0087] Action 201 may be understood to be an optional action. In some examples, instead of capturing the initial expectation about each data distribution from a live network, the data distribution that may have been used when training the original ML model of the first predictive ML model in a laboratory may be used. In order to use that original ML model as a baseline for ascertaining that expectation through the model's lifecycle and to determine when the baseline ML model may need to be retrained. The original ML model may be understood to refer to the very first version of the ML model that may have had high enough performance to be released. As such, it may be considered as a baseline.)
As to claim 10, Vandikas teaches the apparatus according to claim 7, wherein the at least one memory and the instructions, when executed by the at least one processor, further cause the apparatus at least to: provide, to the network entity, a request for retraining the at least one machine learning model; ([0185] In this Action 406, the third node 113 may send, to the fourth node 114, the sixth indication The sixth indication may further indicate the second predictive ML model.)
receive one or more measurement reports that comprise at least one dataset for retraining the at least one machine learning model; and ([0186] For example, the sixth indication may be a tuple wherein the source/destination and corresponding ports, the ML model blueprint, the top features, produced from the feature importance function, and the data distribution of the top features may be assembled e.g., as <source, destination, source port, destination port, mb, top_features, data_distribution>.)
retrain the at least one machine learning model based on the one or more measurement reports. ([0182] In some embodiments, the obtaining in this Action 404 of the second indication may be based on the sent second predictive ML model in Action 402. [0189] By sending the sixth indication in this Action 406, the third node 113 may enable the fourth node 114 to then record the sixth indication, and use this input, to compare the similarity between the data distribution transferred over the network, which may be a vector, with the expected distribution, and train the second predictive ML model to detect any anomalies in the data that may be being transmitted.)
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ELTON S WILLIAMS whose telephone number is (571)272-9933. The examiner can normally be reached 8-4 Mon-Fri.
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, Gary Mui can be reached at (571) 270-1420. 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.
/Elton Williams/ Examiner, Art Unit 2465
Prosecution Insights