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 .
The amendments were received on 6/16/2026. Claims 1-4, 6-14, and 16-20 are pending where claims 1-4, 6-15, and 16-20 were previously presented and claims 5 and 15 were cancelled.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-4, 6-14, and 16-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
With regard to claim 1:
Step 2A, Prong One:
The claim recites the following limitations which are drawn towards an abstract idea:
A neural network optimization method with a preview mechanism, comprising:
determining whether a reference value corresponding to the multiple previewed results exceeds a limitation value; in response to the reference value exceeding the limitation value, stopping previewing the optimization space received in the preview stage (recites mental process steps of evaluation/analysis and comparison to from a determination/judgement (see paragraph 17 for discussion of updating signal to determine whether to continue the search/evaluation or to stop) which relates to evaluating/comparing data sets and deciding that the task is complete, i.e. to stop),
in response to the reference value not exceeding the limitation value, adjusting the optimization space received in the preview stage according to the reference value, to generate multiple adjusted optimization spaces, for collecting as multiple collected optimization spaces (recites mental process steps of evaluating/comparing data sets and decide that more work for the task should continue);
and processing the optimization space received in the view stage according to the updating signal and a setting mode to generate an optimization result (recites mental process steps of evaluation and analysis to form a determination or selection of a particular received data to be the configuration/data of choice).
As seen from above, the identified limitations recite concepts associated with an abstract idea and thus the respective claim recites a judicial exception (see 2106.04(a)) and thus requires further analysis as discussed below.
Step 2A, Prong Two:
The following limitations have been identified as being additional elements as discussed below.
in a preview stage, building an optimization space and obtaining multiple previewed results from the optimization space (recites insignificant extrasolution activity of mere data gathering/receiving information, see MPEP 2106.05(g));
and directly outputting the multiple previewed results to the view stage as the updating signal (recites insignificant extrasolution activity of transmitting information, see MPEP 2106.05(g));
and in a view stage, receiving the optimization space and the updating signal (recites insignificant extrasolution activity of receiving information, see MPEP 2106.05(g)),
wherein in response to the setting mode being a time saving mode, the optimization space received in the view stage is always updated according to the updating signal (recites insignificant extrasolution activity of streaming or real-time updates which recites at a high-level of generality the functionality of transmitting information, see MPEP 2106.05(g));
wherein in response to the setting mode being a computation resource saving mode, the optimization space received in the view stage is updated according to the updating signal only when the reference value exceeds the limitation value (recites insignificant extrasolution activity of transmitting information, see MPEP 2106.05(g));
wherein in response to the setting mode being a thoroughly mode, the multiple collected optimization spaces are thoroughly viewed in the view stage ((recites insignificant extrasolution activity of transmitting information, see MPEP 2106.05(g))).
As seen from the above discussion, the identified limitations did not integrate the judicial exception into a practical application (see MPEP 2106.04(d)). This judicial exception is not integrated into a practical application because the additional elements recite merely gathering/receiving data/information at a high-level of generality.
Step 2B:
Below is the analysis of the claims:
in a preview stage, building an optimization space and obtaining multiple previewed results from the optimization space (recites well-understood, routine, and conventional activity of mere data gathering/receiving information, see MPEP 2106.05(d));
and directly outputting the multiple previewed results to the view stage as the updating signal (recites well-understood, routine, and conventional activity of transmitting information, see MPEP 2106.05(d));
and in a view stage, receiving the optimization space and the updating signal (recites well-understood, routine, and conventional activity of receiving information, see MPEP 2106.05(d));
wherein in response to the setting mode being a time saving mode, the optimization space received in the view stage is always updated according to the updating signal (recites well-understood, routine, and conventional activity of streaming or real-time updates which recites at a high-level of generality the functionality of transmitting information, see MPEP 2106.05(d));
wherein in response to the setting mode being a computation resource saving mode, the optimization space received in the view stage is updated according to the updating signal only when the reference value exceeds the limitation value (recites well-understood, routine, and conventional activity of transmitting information, see MPEP 2106.05(d));
wherein in response to the setting mode being a thoroughly mode, the multiple collected optimization spaces are thoroughly viewed in the view stage ((recites well-understood, routine, and conventional activity of transmitting information, see MPEP 2106.05(d))).
As seen from above, the respective claim elements taken individually do not amount to significantly more than the judicial exception. When taken as a whole (in combination), the claim also does not amount to significantly more than the abstract idea because the additional elements recite merely gathering/receiving data/information at a high-level of generality.
With regard to claim 2, this claim recites wherein the step of obtaining the multiple previewed results from the optimization space comprises: sampling the optimization space to obtain multiple candidate networks (recites mental process steps of selecting particular samples or configurations);
evaluating the multiple candidate networks to obtain multiple evaluated results (recites mental process steps of evaluating/analysis of selected samples/configurations);
and obtaining multiple previewed neural networks from the multiple candidate networks as the multiple previewed results according to the multiple evaluated results (recites mental process step of a judgment/decision on the samples that warrant further consideration).
With regard to claim 3, this claim recites wherein the step of evaluating the multiple candidate networks to obtain the multiple evaluated results comprises: utilizing a quality estimator to estimate quality of the multiple candidate networks for obtaining the multiple evaluated results (recites mental process steps of evaluation/estimating a quality score/value for respective sampled data where the score can be determined via guessing/observations of past behavior/knowledge or computed, possibly via mathematical equations).
With regard to claim 4, this claim recites wherein the step of evaluating the multiple candidate networks to obtain the multiple evaluated results comprises: utilizing a performance estimator to estimate platform performance of the multiple candidate networks for obtaining the multiple evaluated results (recites mental process steps of evaluation/estimating a performance score/value for respective sampled data where the score can be determined via guessing/observations of past behavior/knowledge or computed, possibly via mathematical equations).
With regard to claim 6, this claim recites wherein the reference value is a time for previewing the optimization space received in the preview stage, the limitation value is a predetermined time, and the step of determining whether the reference value corresponding to the multiple previewed results exceeds the limitation value comprises: determining whether the time exceeds the predetermined time (recites mental process steps of comparing/evaluating the amount of time elapsed to some criteria/threshold, similar to a test time limit).
With regard to claim 7, this claim recites wherein the reference value is a metric for previewing the optimization space received in the preview stage, the limitation value is a predetermined criterion, and the step of determining whether the reference value corresponding to the multiple previewed results exceeds the limitation value comprises: determining whether the metric exceeds the predetermined criterion (recites mental process steps of comparing/evaluating based on some metric, e.g. number of times/rounds/iterations to do a task).
With regard to claim 8, this claim recites wherein the step of processing the optimization space received in the view stage according to the updating signal and the setting mode to generate the optimization result comprises: updating the optimization space received in the view stage according to the updating signal, to generate an updated optimization space (recites mental process steps of evaluation and analysis to form a determination or selection of a particular received data to be the configuration/data of choice);
training neural networks in the updated optimization space to obtain a training result (recites generic training steps of a computer element which amounts to usage of a computer as a tool to implement the abstract idea, see MPEP 2106.05(f));
optimizing the neural networks in the updated optimization space according to the training result, to generate optimized neural networks (recites training/optimizing the respective model which can amount to evaluating/testing the neural network and amounts to usage of a computer as a tool to implement the abstract idea by performing the judicial exception on a computer);
and fine-tuning the optimized neural networks to generate the optimization result (recites training/optimizing the respective model which can amount to evaluating/testing the neural network, possible a trial run to receive feedback or with other training/testing data and amounts to usage of a computer as a tool to implement the abstract idea by performing the judicial exception on a computer).
With regard to claim 9, this claim recites wherein the step of processing the optimization space received in the view stage according to the updating signal and the setting mode comprises: stopping viewing the optimization space received in the view stage according to the updating signal (recites mental process step of evaluation/comparison to form a judgement to stop doing something).
With regard to claim 10, this claim recites wherein the neural network optimization method is applied to a neural architecture search (NAS) (recites field of use limitations describing a particular, or preferred, technique, see MPEP 2106.05(h)).
With regard to claims 11-14 and 16-20, these claims are substantially similar to claims 1-4 and 6-10 and are rejected for similar reasons as discussed above.
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 (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 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-5, 7-15, and 17-20 are rejected under 35 U.S.C. 103 as being unpatentable over Akhauri et al [US 2022/0108054 A1] in view of Chang et al [US 2023/0359885 A1], Galatzer et al [US 2018/0107740 A1], and Li et al [US 2010/0312782 A1].
With regard to claim 1, Akhauri teaches a neural network optimization method with a preview mechanism, comprising: in a preview stage, building an optimization space and obtaining multiple previewed results from the optimization space (see paragraphs [0020], [0034], [0035], and [0060]-[0061]; the system can utilize an optimized space and be able to receive multiple candidate/previewed results of a respective subspace);
and in a view stage, receiving the optimization space
Akhauri does not appear to explicitly teach:
determining whether a reference value corresponding to the multiple previewed results exceeds a limitation value;
in response to the reference value exceeding the limitation value, stopping previewing the optimization space received in the preview stage, and directly outputting the multiple previewed results to the view stage as the updating signal;
in response to the reference value not exceeding the limitation value, adjusting the optimization space received in the preview stage according to the reference value, to generate multiple adjusted optimization spaces, for collecting as multiple collected optimizations spaces;
and in a view stage, receiving the optimization space and the updating signal, and processing the optimization space received in the view stage according to the updating signal and a setting mode to generate an optimization result;
wherein in response to the setting mode being a time saving mode, the optimization space received in the view stage is always updated according to the updating signal;
wherein in response to the setting mode being a computation resource saving mode, the optimization space received in the view stage is updated according to the updating signal only when the reference value exceeds the limitation value;
wherein in response to the setting mode being a thoroughly mode, the multiple collected optimization spaces are thoroughly viewed in the view stage.
Chang teaches determining whether a reference value corresponding to the multiple previewed results exceeds a limitation value (see paragraph [0060]; the system can utilize a reference value for determining how many iterations to perform of the updating machine learning model).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify the training/evaluation process of various models/algorithms of Akhauri by utilizing additional parameters such as number of iterations/epochs as a reference value as taught by Chang in order to allow the system to evaluate the various models/algorithms while also considering a reference value or some sufficient number of iterations so that the system doesn’t unnecessarily bog itself down with algorithms/models that won’t converge without extensive training thus allowing for a fair evaluation of the respective models with respect to their performance after some sufficient amount of time training/updating.
Akhauri in view of Chang do not appear to explicitly teach in response to the reference value exceeding the limitation value, stopping previewing the optimization space received in the preview stage, and directly outputting the multiple previewed results to the view stage as the updating signal; in response to the reference value not exceeding the limitation value, adjusting the optimization space received in the preview stage according to the reference value, to generate multiple adjusted optimization spaces, for collecting as multiple collected optimizations spaces (see Akhauri, paragraphs [0020], [0037]-[0038], and [0064]-[0065]; see Chang, paragraph [0060]; the system can utilize the reference value to allow for sufficient iterations to occur with respect to the relative value to determine when to stop previewing/evaluating respective candidates as well as be able to generate an update signal based on evaluation to reference values and process an action accordingly including identifying whether the preview stage is complete and, if so, to use the determined/selected candidate controller/result and associated parameters or to continue the searching);
and in a view stage, receiving the optimization space and the updating signal, and processing the optimization space received in the view stage according to the updating signal
Akhauri in view of Chang do not appear to explicitly teach:
and in a view stage, receiving the optimization space and the updating signal, and processing the optimization space received in the view stage according to the updating signal and a setting mode to generate an optimization result;
wherein in response to the setting mode being a time saving mode, the optimization space received in the view stage is always updated according to the updating signal;
wherein in response to the setting mode being a computation resource saving mode, the optimization space received in the view stage is updated according to the updating signal only when the reference value exceeds the limitation value;
wherein in response to the setting mode being a thoroughly mode, the multiple collected optimization spaces are thoroughly viewed in the view stage.
Galatzer teaches a setting mode to generate an optimization result; wherein in response to the setting mode being a time saving mode,
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify the training/evaluation process of various models/algorithms of Akhauri in view of Chang by providing means to allow the user of the system to provide settings/preferences for how the user(s) want the system to behave with respect to result collections as taught by Galatzer in order to increase system functionality and versatility by including the ability to provide results as they are collected/gathered or wait for all the results to be retrieved and illustrate the complete results which can be tailored/configured based on user preferences.
Akhauri in view of Chang and Galatzer teach in a view stage, receiving the optimization space and the updating signal, and processing the optimization space received in the view stage according to the updating signal and a setting mode to generate an optimization result (see Akhauri, paragraph [0062]; see Chang, paragraph [0060]; see Galatzer, paragraph [0095]; the system can utilize the search space/optimization space and be able to process that space with respect to the selected result and be optimized for that entire space including having performances or setting modes);
wherein in response to the setting mode being a time saving mode, the optimization space received in the view stage is always updated according to the updating signal; wherein in response to the setting mode being a computation resource saving mode, the optimization space received in the view stage is updated according to the updating signal only when the reference value exceeds the limitation value (see Galatzer, paragraph [0095]; see Akhauri paragraphs [0063] and [0037]; the system can have the resulting optimization or search space updated with new results are received to save time by being able to display results to the user immediately or by waiting for the complete result set first before displaying/outputting results to save resources having to constantly re-transmit results to the user/client and from having to perform merging/sorting/ordering operations of the previously received results with the newly received results).
Akhauri in view of Chang and Galatzer do not appear to explicitly teach:
wherein in response to the setting mode being a thoroughly mode, the multiple collected optimization spaces are thoroughly viewed in the view stage.
Li teaches the multiple collected optimization spaces are thoroughly viewed in the view stage (see Figure 2 and paragraphs [0022] and [0024]; the system can present the multiple collected spaces/results in separate spaces so that they can be thoroughly viewed).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify the training/evaluation process of various models/algorithms of Akhauri in view of Chang and Galatzer by being able to not only display results as a complete set or intermediate set but also be able to allow users to decide on if they want a single listing or separated listings as taught by Li in order to change how results or candidates are presented to the user so that instead of a single list of results the system can be able to have multiple lists based on different subspaces so that the user can readily or thoroughly review results including best results from the various different subspaces thus allowing for quicker understanding of how well different results/candidates performed for the various search spaces/subspaces.
Akhauri in view of Chang, Galatzer, and Li teach wherein in response to the setting mode being a thoroughly mode, the multiple collected optimization spaces are thoroughly viewed in the view stage (see Li, Figure 2 and paragraphs [0022] and [0024]; see Akhauri, paragraph [0062]; see Chang, paragraph [0060]; see Galatzer, paragraph [0095]; the system’s preferences can also indicate the user’s preferred output format including if they want to be able to thoroughly review the various results by the different subspaces).
With regard to claim 2, Akhauri in view of Chang, Galatzer, and Li teach wherein the step of obtaining the multiple previewed results from the optimization space comprises: sampling the optimization space to obtain multiple candidate networks; evaluating the multiple candidate networks to obtain multiple evaluated results; and obtaining multiple previewed neural networks from the multiple candidate networks as the multiple previewed results according to the multiple evaluated results (see Akhauri, paragraphs [0020], [0034], [0035], and [0060]; the system can sample the optimization space and evaluate multiple candidate networks to determine candidate controllers/networks that can be ranked/scored).
With regard to claim 3, Akhauri in view of Chang, Galatzer, and Li teach wherein the step of evaluating the multiple candidate networks to obtain the multiple evaluated results comprises: utilizing a quality estimator to estimate quality of the multiple candidate networks for obtaining the multiple evaluated results (see Akhauri, paragraphs [0038] and [0064]-[0065]; the system can utilize some methodology to estimate expected quality of the respective candidate network).
With regard to claim 4, Akhauri in view of Chang, Galatzer, and Li teach wherein the step of evaluating the multiple candidate networks to obtain the multiple evaluated results comprises: utilizing a performance estimator to estimate platform performance of the multiple candidate networks for obtaining the multiple evaluated results (see Akhauri, paragraph [0037]; the system allows for the evaluation of the respective candidates with respect to performance within the respective search subspace).
With regard to claim 7, Akhauri in view of Chang, Galatzer, and Li teach wherein the reference value is a metric for previewing the optimization space received in the preview stage, the limitation value is a predetermined criterion, and the step of determining whether the reference value corresponding to the multiple previewed results exceeds the limitation value comprises: determining whether the metric exceeds the predetermined criterion (see Chang, paragraph [0060]; see Akhauri, paragraph [0037]; the system allows for some predetermined criterion to be utilized as the reference value with means to determine if the metric value has been exceeded).
With regard to claim 8, Akhauri in view of Chang, Galatzer, and Li teach wherein the step of processing the optimization space received in the view stage according to the updating signal to generate the optimization result comprises: updating the optimization space received in the view stage according to the updating signal and setting mode, to generate an updated optimization space; training neural networks in the updated optimization space to obtain a training result; optimizing the neural networks in the updated optimization space according to the training result, to generate optimized neural networks; and fine-tuning the optimized neural networks to generate the optimization result (see Akhauri, paragraph [0036]-[0037] and [0060]-[0062]; see Chang, paragraph [0060]; the system can utilize the learned parameters associated with each respective optimized design of the respective candidates for scoring with respect to additional subspaces including re-training/tuning to determine their overall results in order to find the controller that most frequently performances best for the various optimization spaces).
With regard to claim 9, Akhauri in view of Chang, Galatzer, and Li teach wherein the step of processing the optimization space received in the view stage according to the updating signal and the setting mode comprises: stopping viewing the optimization space received in the view stage according to the updating signal (see Akhauri, paragraphs [0062] and [0064]; the system can determine via an updating signal whether the system should stop the viewing/evaluating of the various candidates since a candidate has been found that meets various performance and quality criteria).
With regard to claim 10, Akhauri in view of Chang, Galatzer, and Li teach wherein the neural network optimization method is applied to a neural architecture search (NAS) (see Akhauri, paragraphs [0034], [0041], and [0103]; see Chang, paragraph [0047]; NAS is used).
With regard to claim 11, this claim is substantially similar to claim 1 and is rejected for similar reasons as discussed above.
With regard to claims 12-14 and 17-20, these claims are substantially similar to claims 2-4 and 7-10 respectively and are rejected for similar reasons as discussed above.
Claims 6 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Akhauri et al [US 2022/0108054 A1] in view of Chang et al [US 2023/0359885 A1], Galatzer et al [US 2018/0107740 A1], and Li et al [US 2010/0312782 A1] in further view of Benyahia et al [US 2020/0104688 A1].
With regard to claim 6. Akhauri in view of Chang, Galatzer, and Li teach all the claim limitations of claims 1 and 5 as discussed above.
Akhauri in view of Chang, Galatzer, and Li do not appear to explicitly teach wherein the reference value is a time for previewing the optimization space received in the preview stage, the limitation value is a predetermined time, and the step of determining whether the reference value corresponding to the multiple previewed results exceeds the limitation value comprises: determining whether the time exceeds the predetermined time.
Benyahia teaches wherein the reference value is a time for previewing the optimization space received in the preview stage, the limitation value is a predetermined time, and the step of determining whether the reference value corresponding to the multiple previewed results exceeds the limitation value comprises: determining whether the time exceeds the predetermined time (see paragraph [0109]; the system can train the respective candidate models based on a reference value associated with a time threshold with training continuing until the respective time threshold is met).
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify the training/evaluation process of various models/algorithms of Akhauri in view of Chang by utilizing a time threshold as means for determining how long training should occur as taught by Benyahia in order to allow the system to evaluate the various models/algorithms within a time period thus helping to ensure that the system doesn’t unnecessarily bog itself down with algorithms/models that take a long time even when there is a large quantity of training data/batches that can be used thus helping to see how the respective models perform equally since they all have the same amount of time training.
With regard to claim 16, this claim is substantially similar to claim 6 and is rejected for similar reasons as discussed above.
Examiner Comments
In view of applicant’s amendments and arguments, the Examiner noticed concepts similar to those that applicant is claiming that could be expanded upon to help differentiate the claims from the prior art of record. The concepts relate to paragraphs [0020]-[0021] of the originally filed specification that discusses the adaptor stage and various operations that occur there. Although the claims discuss some of the operations, clarifying details regarding the various components and which components are doing the various operations/functions can help provide details that could differentiate the claims from the prior art. Additionally, or alternatively, with regards to the new limitations, instead of only ‘in response to’ limitations, have an additional limitation that is a determining or detecting step that identifies a setting mode that would actively make the setting mode be one of the three listed setting modes. Also, expanding on the process of generate adjusted optimization spaces discussed in paragraph [0021] relating to the quality and steps for enlarging or lowering the optimization space could also help provide clarifying details to differentiate the claims from the prior art of record. The Examiner notes that the above noted concepts would require further search and consideration.
Response to Arguments
Applicant's arguments (see the second to last paragraph on page 9 through the fourth paragraph on page 11) have been fully considered but they are not persuasive. The applicant argues that the amended limitations improve the overall efficiency of neural network optimization by reducing unnecessary processing of inferior optimization spaces and/or conserving computational resources which integrates the abstract idea into a practical application. The Examiner respectfully disagrees. The claims have been amended to include new limitations associated with the setting mode; however, with respect to claim 1, the limitation recites that the system processes the optimization space according to ‘the updating signal and a setting mode to generate an optimization result’ which relates to the judicial exception. Per MPEP 2106.05(a), the judicial exception alone cannot provide the improvement and also an important consideration is the extent that the claim covers a particular solution. With regard to the “wherein in response to the setting mode being a…” limitations with the time saving mode, computation resource saving mode, or the thoroughly mode, the Examiner notes that these are additional elements that are conditional under the broadest reasonable interpretation of the claim, i.e. the claim only requires a setting mode but is not restrictive to the listed modes. Additionally, the respective limitations recite different means of collecting/receiving information such as continuously (i.e. streaming) results as they arrive (i.e. time saving mode) or batching results in a complete set (computation resource saving). The thoroughly mode is recited at a high-level of generalization that can relate to multiple reasonable interpretations including presenting/displaying/transmitting extensive amounts of information. As such, due to the conditional nature of those additional elements and their broad recitations, the applicant’s arguments are not persuasive.
Applicant’s arguments (see the last paragraph on page 11 through the last paragraph on page 13) with respect to the rejection(s) of claim(s) under Akhauri and Chang have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Galatzer and Li. The applicant amended the claims to incorporate new limitations that required further search and consideration. As noted in the updated 35 USC 103 rejections above, new references were found that, when combined, would teach or fairly suggest the claim limitations as recited.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Abdelfattah et al [US 2021/0019599 A1] teaches at paragraph [0065] that the system can provide a display screen to display the results of the NAS as well as teaches in Figure 1 having a reference value to determine when to stop searching.
Singh et al [US 2020/0302270 A1] teaches at Figure 4 and paragraphs [0042]-[0044] a neural architecture search that is based on performance budgeting variables including determining if there is time left or compute available to generate a new candidate architecture.
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 nonprovisional extension fee (37 CFR 1.17(a)) 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 mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MARC S SOMERS whose telephone number is (571)270-3567. The examiner can normally be reached M-F 11-8 EST.
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, Ann Lo can be reached at 5712729767. 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.
/MARC S SOMERS/Primary Examiner, Art Unit 2159 7/17/2026
Prosecution Insights