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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 04/24/2026 has been entered.
Response to Arguments
The applicant's arguments/remarks, filed 04/24/2026, see page 6-8, with respect to 35 U.S.C 102 and 103 rejections of Claims 1, 2, 5-11 and 13-18 have been fully considered but are moot in view of the new ground(s) of rejection. The arguments/remarks are essentially directed towards the newly introduced limitations and they are addressed in this Office Action, below.
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.
Claim(s) 1, 2, 5, 6, 8, 11, 13, 14 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kou et al. (US PG PUB 20230229890), hereinafter "Kou", in views of Tullberg et al. (US PG PUB 20220078637), hereinafter "Tullberg".
Regarding Claims 1 and 11, Kou discloses:
(Claim 1) A method of machine learning performed by a wireless transmit receive unit (WTRU) (i.e. system 100 [i.e. a wireless transmit receive unit WTRU] comprised of a plurality of networked computers, e.g. computing devices 105-1, 105-n and second computing system 130, that may communicate each other wirelessly, may implement training and updating of neural networks [i.e. performing machine learning]; Note that the system 100 is a part of [i.e. one of a unit] of the larger network 145) (Fig. 1, ¶ 0024 and ¶ 0046 – 0049), the method comprising:
(Claim 11) A wireless transmit receive unit (WTRU) (i.e. system 100 [i.e. a wireless transmit receive unit WTRU] comprised of a plurality of networked computers, e.g. computing devices 105-1, 105-n and second computing system 130, that may communicate each other wirelessly; Note that the system 100 is a part of [i.e. one of a unit] of the larger network 145) (Fig. 1, ¶ 0024 and ¶ 0046 – 0049) being configured to:
Implementing/implement, by a first machine learning (ML) module, a first ML model using input data to generate first prediction data (i.e. smart box edge computing device [i.e. a first machine learning ML module] may implement 1st NN model 110 [i.e. a first ML model] to generate results [i.e. first prediction data] using input data, e.g. data collected by one or more sensors) (110 – Fig. 1, ¶ 0037, ¶ 0046 and ¶ 0052),
wherein the first ML model is a production model (i.e. the 1st NN model 110 [i.e. the first ML model] is used for generating/producing results [i.e. the first ML model is a production model]) (¶ 0011 and ¶ 0046);
transmitting/transmit, to a second ML module at a network node, the input data and the first prediction data (i.e. the method/system, e.g. first computing device 105 of the system 100 [i.e. a wireless transmit receive unit WTRU], may send/transmit at least some of the set of results [i.e. the first prediction data] and the corresponding input data [i.e. the input data] to local smart box [i.e. a second ML module] of the second computing system 130 [i.e. a network node]) (130 & 135 – Fig. 1, 205 & 210 – Fig. 2, ¶ 0011 and ¶ 0052),
wherein the second ML module includes a second ML model (i.e. the local smart box [i.e. the second ML module], includes a second 2nd Neural network model 135, e.g. heavyweight model [i.e. a second ML model]) (130 & 135 – Fig. 1, 205 & 210 – Fig. 2, ¶ 0011, ¶ 0037, ¶ 0043 and ¶ 0052); and
(Claim 11) execute the first ML model (i.e. the computing device may execute the lightweight model [i.e. the first ML model]) (110 – Fig. 1, ¶ 0037, ¶ 0046 and ¶ 0052).
However, Kou does not explicitly disclose:
receiving/receive, by the first ML module, from the second ML module of the network node, an accuracy metric transmitted by the second ML module and based on a comparison of the transmitted first prediction data of the first ML model and second prediction data obtained from the second ML model; and updating/update, by the first ML module, the first ML model based on the received accuracy metric and an accuracy condition.
On the other hand, in the same field of endeavor, Tullberg teaches:
receiving/receive, by the first ML module, from the second ML module of the network node, an accuracy metric transmitted by the second ML module (i.e. the wireless device 120 comprised of first instance of machine learning model [i.e. the first ML module] may receive information relating to the updated one or more parameters [i.e. an accuracy metric; Note that the updated one or more parameters are for reducing difference between the first and second instance of the machine learning models; In other words they are for enhancing accuracy; Therefore, the one or more parameters are accuracy metric] transmitted by the network node 110, 130, etc. comprised of second instance of machine learning model [i.e. the second ML module]) (205 – Fig. 2, ¶ 0106 and ¶ 0109) and
based on a comparison of the transmitted first prediction data of the first ML model and second prediction data obtained from the second ML model (i.e. the one or more parameters [i.e. the accuracy metric] is obtained based on comparison between a prediction of the operation obtained by the second instance [i.e. second prediction data obtained from the second ML model] and the prediction of the first instance of the machine learning model [i.e. first prediction data of the first ML model]) (¶ 0106 - 0107); and
updating/update, by the first ML module, the first ML model based on the received accuracy metric and an accuracy condition (i.e. the first instance of the machine learning model is updated [i.e. updating/update, by the first ML module, the first ML model] based on the one or more parameters [i.e. the received accuracy metric] and indication of the model difference being above threshold [i.e. an accuracy condition] in order to improve the predictions) (¶ 0107 – 0109).
Therefore, 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 method/system of Kou to include for receiving/receive, by the first ML module, from the second ML module of the network node, an accuracy metric transmitted by the second ML module and based on a comparison of the transmitted first prediction data of the first ML model and second prediction data obtained from the second ML model; and updating/update, by the first ML module, the first ML model based on the received accuracy metric and an accuracy condition as taught by Tullberg so that machine learning module of the WTRU may be updated based on the model differences between the current model at the WTRU and updated models of the network node (¶ 0107 – 0109).
Regarding Claim 2, Kou and Tullberg disclose, in particular Kou teaches:
executing, by the first ML module, the first ML model (i.e. smart box edge computing device [i.e. a first machine learning ML module] may execute 1st NN model 110 [i.e. a first ML model] to generate results) (110 – Fig. 1, ¶ 0037, ¶ 0046 and ¶ 0052).
Regarding Claims 5 and 13, Kou and Tullberg disclose, in particular Kou teaches:
Wherein the input data are received from the WTRU (i.e. first computing device 105 [i.e. a wireless transmit receive unit (WTRU)] which may be a AI camera captures the input data) (105 & 110 – Fig. 1, ¶ 0024, ¶ 0040 and ¶ 0046).
Regarding Claims 6 and 14, Kou and Tullberg disclose, in particular Kou teaches:
wherein the second MVL model has any of: (1) a greater accuracy metric than the first ML model for a predetermined validation data set, (2) a greater number of floating-point operations, and (3) a greater memory size (i.e. The heavyweight model is likely to be more complex—thereby using more computing resources (memory, computing time, energy, processing power, etc.) [i.e. a greater memory size], but it achieves higher accuracy than the lightweight model [i.e. a greater accuracy metric than the first ML model for a predetermined validation data set]) (¶ 0037).
Regarding Claims 8 and 16, Kou and Tullberg disclose, in particular Kou teaches:
Generating/generate a dataset, wherein the dataset comprises input data associated with at least a second prediction data of the second prediction data generated by the second ML module (i.e. a training dataset may be formed by using the collected input data [i.e. comprises input data associated with at least a second prediction data] as input data and the corresponding results from the second neural network as ground truth results [i.e. a second prediction data of the second prediction data generated by the second ML module]) (¶ 0005 - 0006).
Claim(s) 7 and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kou in views of Tullberg as applied to claims 1 and 11 above, and further in view of Ghanta et al. (US PG PUB 20200034665), hereinafter "Ghanta".
Regarding Claims 7 and 15, Kou and Tullberg disclose all the features with respect to Claim 1 as described above.
However, the combination of Kou and Tullberg does not explicitly disclose:
Wherein the first ML model is updated by selecting a third ML model among one or more candidate ML models.
On the other hand, in the same field of endeavor, Ghanta teaches:
Wherein the first ML model is updated by selecting a third ML model among one or more candidate ML models (i.e. the action module 312 is configured to trigger an action associated with the first machine learning algorithm in response to the predicted suitability of the first machine learning algorithm/model for analyzing the inference data set not satisfying a predetermined suitability threshold; the action may include selecting a machine learning model [i.e. a third ML model] from a plurality of machine learning models [i.e. one or more candidate ML models]) (¶ 0071 and ¶ 0101).
Therefore, 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 method/system of Kou and Tullberg to include the feature wherein the first ML model is updated by selecting a third ML model among one or more candidate ML models as taught by Ghanta so that a best suited machine learning model may be activated in response to the predicted suitability of the first machine learning algorithm/model for analyzing the inference data set not satisfying a predetermined suitability threshold (¶ 0071 and ¶ 0101).
Claim(s) 9 and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kou in views of Tullberg as applied to claims 8 and 16 above, and further in view of Jung (US PG PUB 20210056412), hereinafter "Jung".
Regarding Claims 9 and 17, Kou and Tullberg disclose all the features with respect to Claim 8 as described above.
However, the combination of Kou and Tullberg does not explicitly disclose:
wherein the at least second prediction data is associated with a confidence score, and wherein generating the dataset further comprises adding to the dataset the at least second prediction data, based on the confidence score associated with the at least second predictions data.
On the other hand, in the same field of endeavor, Jung teaches:
wherein the at least second prediction data is associated with a confidence score (i.e. data generated by neural network [i.e. the at least second prediction data] is associated with corresponding confidence level [i.e. confidence score]) (Abstract, ¶ 0025 and ¶ 0100), and
wherein generating the dataset further comprises adding to the dataset the at least second prediction data, based on the confidence score associated with the at least second predictions data (i.e. training dataset may be generated by labeling the subset of candidate data from among the subset of candidate data in accordance with a confidence level label based on the confidence conditions) (Abstract, ¶ 0025 and ¶ 01000).
Therefore, 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 method/system of Kou and Tullberg to include the feature wherein the at least second prediction data is associated with a confidence score, and wherein generating the dataset further comprises adding to the dataset the at least second prediction data, based on the confidence score associated with the at least second predictions data as taught by Jung so that a training dataset may be generated in accordance with the confidence levels associated with the candidate data) (Abstract, ¶ 0025 and ¶ 01000).
Claim(s) 10 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kuo in views of Tullberg as applied to claims 8 and 16 above, and further in view of Sharma et al. (US PG PUB 20220391312), hereinafter "Sharma".
Regarding Claims 10 and 18, Kuo and Tullberg disclose all the features with respect to Claim 8 as described above.
However, the combination of Kuo and Tullberg does not explicitly disclose:
Wherein the first ML model is retrained by the first ML module, using the generated dataset.
On the other hand, in the same field of endeavor, Sharma teaches:
Wherein the first ML model is retrained by the first ML module, using the generated dataset (i.e. ML prediction model [i.e. the first ML model] may be retrained by RA platform [i.e. the first ML module] using the updated training set [i.e. generated dataset]) (Fig. 1, Fig. 2, ¶ 0021 and ¶ 0060 - 0061).
Therefore, 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 method/system of Kou and Tullberg to include the feature wherein the first ML model is retrained by the first ML module, using the generated dataset as taught by Sharma so that machine learning model may be retrained based on detection of the inaccuracy in the output of the machine learning model Fig. 1, Fig. 2, ¶ 0021 and ¶ 0060 - 0061).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SOE MIN HLAING whose telephone number is (303)297-4282. The examiner can normally be reached Monday-Friday 9AM - 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, Christopher Parry can be reached at 571-272-8328. 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.
/Soe Hlaing/ Primary Examiner, Art Unit 2451
Prosecution Insights