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 .
Examiner’s Note
Regarding 35 USC § 103 rejection, the rejection made in the previous action has been withdrawn
in light of the instant amendments to the claims.
Response to argument
Applicant's arguments filed 06/16/2025 ("Arguments/Remarks") have been fully considered but they are not persuasive.
Argument – 1: (page: 13) applicant contends: “As described in paragraph [0008] of Applicant’s specification, “Embodiments of the present invention provide a method for controlling multiple parallel-operating machine learning pipelines where feature evaluation, model selection, and confidence scoring is performed in reduced time and with reduced computational resources. Embodiments of the present invention recognize that the utilization of stop criteria in machine learning pipelines produce high confidence predictions with reduced computational processing, features and subsequent model generations. Embodiments of the present invention generate conclusions from the results of an ensemble of maintained pipelines, while concurrently, allowing incomplete solutions from the maintained pipelines running in parallel. In this embodiment, the present invention identifies results sooner without requiring full processing of all the pipelines”.”
Regarding argument – 1, the Examiner respectfully disagrees with Applicant’s assertion that utilization of stop criteria in machine learning pipelines to produce high confidence predictions or identifies results sooner without requiring full processing of all the pipelines improve the functioning of a computer or technical field. It lacks sufficient details required to support a conclusion that the claim recites a technological improvement. The claims require sufficient technical details to how the stop criteria are implemented or how they lead to claimed improvements. Without a clear description of the underlying mechanism or how the stop criteria function within the machine learning pipeline, the claims fail to provide sufficient support to demonstrate an improvement. To determine if the disclosure provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement. The specification need not explicitly set forth the improvement, but it must describe the invention such that the improvement would be apparent to one of ordinary skill in the art. Conversely, if the specification explicitly sets forth an improvement but in a conclusory manner (i.e., a bare assertion of an improvement without the detail necessary to be apparent to a person of ordinary skill in the art) the examiner should not determine the claim improves technology. MPEP 2106.04(d)(1).
Argument – 2: (page: 14) applicant contends “These limitations recite additional elements that are beyond the asserted judicial exception because they provide meaningful limitations on the alleged abstract idea. Thus, these elements add meaningful limitations that control multiple parallel-operating machine learning pipelines where feature evaluation, model selection, and confidence scoring is performed in reduced time and with reduced computational resources, which prior to the Applicant’s disclosure, was not known/possible.”
Regarding argument – 2, the Examiner respectfully disagrees with Applicant’s assertion that the additional limitations go beyond mere assertion of a judicial exception or integrates the judicial exception into practical application. The additional limitations presented in conclusory manner and lacks sufficient technical details to demonstrate integrating the judicial exception into practical application. Without specifics on implementation, this remains a high-level assertion rather than a clear technological improvement. The Examiner also notes that limitations that are identified by the Applicant as additional elements to establish improvement are part of the recited abstract idea itself. As such, they cannot contribute to integrating the abstract idea into practical application. The claimed improvement has to come from the additional element not from abstract idea.
Argument – 3: (page: 15) applicant contends “ ““Moreover, “A claim reciting a judicial exception is not directed to the judicial exception if it also recites additional elements demonstrating that the claim as a whole integrates the exception into a practical application. One way to demonstrate such integration is when the claimed invention improves the functioning of a computer or improves another technology or technical field”. (MPEP 2106.04(d)(1)).””
Regarding argument – 3, the Examiner notes that, as clearly explained above, the claims require sufficient technical details to how the stop criteria are implemented or how they lead to claimed improvements. Without a clear description of the underlying mechanism or how the stop criteria function within the machine learning pipeline, the claims fail to provide sufficient support to demonstrate an improvement. To determine if the disclosure provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement.
Argument – 4: (page: 17) applicant contends “Therefore, Applicant respectfully submits that the pending claims include specific unconventional approaches to control multiple parallel-operating machine learning pipelines where feature evaluation, model selection, and confidence scoring is performed in reduced time and with reduced computational resources. Because the combination of limitations in the pending claims operate in a nonconventional and non-generic way to control multiple parallel- operating machine learning pipelines where feature evaluation, model selection, and confidence scoring is performed in reduced time and with reduced computational resources, the pending claims are not directed to an abstract idea, and are therefore patent-eligible under 35 U.S.C. § 101.”
Regarding argument – 4, the Examiner notes that, the claims lack sufficient details required to support a conclusion that the claim recites a technological improvement. The claims require sufficient technical details to how the additional limitations lead to claimed improvements. Without a clear description of the underlying mechanism or how the stop criteria function within the machine learning pipeline, the claims fail to provide sufficient support to demonstrate an improvement. To determine if the disclosure provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement. The specification need not explicitly set forth the improvement, but it must describe the invention such that the improvement would be apparent to one of ordinary skill in the art. Conversely, if the specification explicitly sets forth an improvement but in a conclusory manner (i.e., a bare assertion of an improvement without the detail necessary to be apparent to a person of ordinary skill in the art) the examiner should not determine the claim improves technology. MPEP 2106.04(d)(1).
Applicant’s argument (regarding 35 USC § 103 rejection: pg. 17 – 22) with respect to amended claim(s): “wherein each neural network in the plurality of neural networks has a distinctive stop criteria specific to a neural network structure respectively associated with each neural network in the plurality of neural networks, wherein weights of a previously trained model that failed to exceed a confidence threshold are utilized. …the respective stop criteria to increase a prediction duration to allow for computationally intensive neural networks to finish predictions to contribute to the aggregated prediction” have been fully considered and are persuasive. The 35 U.S.C. § 103 rejection has been withdrawn.
As to the remaining dependent claims, applicant argue that they are allowable due to their respective direct and indirect dependencies upon one of the aforementioned Independent claims. The examiner respectfully disagrees, independent claims were not allowable as stated in the paragraph above in this “Response to Arguments” section in this office action.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim(s) 1, 8 and 15 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, or for pre-AIA the applicant regards as the invention.
Claim(s) 1, 8 and 15 recites the limitation “wherein each neural network in the plurality of neural networks has a distinctive stop criteria specific to a neural network structure respectively associated with each neural network in the plurality of neural networks, wherein weights of a previously trained model that failed to exceed a confidence threshold are utilized;”. There is insufficient antecedent basis for the limitation in the claim, which arises from the ambiguity of reference. The term lacks clarity or specific reference to how the weights of a previously trained model that failed to exceed a confidence threshold are utilized in subsequent steps of the method. Without identifying how the weights influence or contributes to generate an aggregated prediction or adjusting stop criteria neural networks, the claim lacks the clarity needed to inform a person of ordinary skill in the art of its scope with reasonable certainty. The claim could be amended to explicitly reflect the relevant teachings from ¶[0022], to particularly explain how the weight is utilized in a subsequent retraining iteration. For examination purposes, the Examiner interprets the weights of a previously trained model, that failed to exceed a confidence threshold, are utilized in a subsequent retraining iteration.
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.
Claim(s) 1 - 20 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
In step 1, of the 101-analysis set forth in the MPEP 2106, the examiner has determined
that the following limitations recite a process that, under the broadest reasonable interpretation, falls within one or more statutory categories (process).
In step 2A prong 1, Determine whether a claim recites an abstract idea by (1) identifying the specific limitation(s) in the claim under examination that the examiner believes recites (i.e. sets forth or describes) an abstract idea, and (2) determining whether the identified limitation(s) fall within at least one of the groupings of abstract ideas.
Regarding claim 1,
determining, …, one or more neural networks within the plurality of neural networks to incorporate a plurality of determined features from the streaming data, comprising;
(i.e.: the broadest reasonable interpretation, the claim recites abstract idea: mental process. Involves a step of making an observation regarding the features extracted from streaming data. Based on this observation, a decision is then made about which neural networks should incorporate these features for further processing or analysis).
calculating, …, a reduction in entropy from a transformation of the dataset based on an information gain of each feature in the plurality of determined features
(i.e.: the broadest reasonable interpretation, the claim recites abstract idea: mental process of evaluating a dataset to reduce its complexity or randomness and it can practically be performed in the human mind, or by a human using pen and paper as a physical aid. For example: consider a dataset with two features, A and B, representing measurements from 10 samples. Initially, the data points are scattered, showing some relationship between A and B but with noticeable variability. To simplify the dataset, one can create a new feature, C, which combines A and B through a specific transformation, such as their ratio or sum. This transformation reduces the dataset’s complexity by consolidating the information from A and B into a single, more coherent feature.)
generating …, an aggregated prediction utilizing each neural network in the plurality of neural networks, in parallel, in the determined plurality of neural networks subject to the respective stop criteria,
(i.e.: the broadest reasonable interpretation, the claim recites abstract idea: mental process of evaluating predictions from machine learning models to create a combined result, which can be reasonably be performed in one’s mind with the aid of pencil and paper.)
calculating, …, a confidence value for the aggregated prediction
(i.e.: the broadest reasonable interpretation, the claim recites abstract idea: mental process of evaluating predictions from machine learning models to calculate a confidence value.)
responsive to the calculated confidence value for the aggregated prediction not reaching a confidence threshold, adjusting, …,the respective stop criteria to increase a prediction duration to allow for computationally intensive neural networks to finish predictions to contribute to the aggregated prediction.
(i.e.: the broadest reasonable interpretation, the claim recites abstract idea: mental process of calculating a confidence value, comparing it to a threshold, and adjusting a parameter (stop criteria) for these neural networks who failed to reach a confidence threshold.)
If the claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process, but for the recitation of generic computer components, then it falls within the mental process. Accordingly, the claim recites an abstract idea.
Step 2A Prone: Two: This part of the eligibility analysis evaluates whether the claim as a whole integrates the recited judicial exception into a practical application.
As evaluated below:
• The preamble is deemed insufficient to transform the judicial exception to a patentable
invention to a patentable invention because the preamble generally links the use of a
judicial exception to a particular technological environment or field of use, see MPEP
2106.0S(h).
by one or more computer processors, (i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation which does not amount to more than a recitation of the words "apply it" (or an equivalent), such as mere instructions to implement an abstract idea on a computer2106.05(f).)
responsive to streaming data, continuously initiating, …, a plurality of parallel-operating machine learning pipelines with the streaming data,
(i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation directed to mere data gathering as deemed insufficient to transform the judicial exception because claimed elements are considered insignificant extra-solution activity, See MPEP (2106.05(g))).
wherein the plurality of parallel-operating machine learning pipelines comprises a plurality of neural networks
(i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation which simply links the judicial exception to a field of use and/or technology environment, see MPEP 2106.05(h)).
wherein each neural network in the plurality of neural networks has a distinctive stop criteria specific to a neural network structure respectively associated with each neural network in the plurality of neural networks
(i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation which simply links the judicial exception to a field of use and/or technology environment, see MPEP 2106.05(h)).
wherein weights of a previously trained model that failed to exceed a confidence threshold are utilized
(i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation which simply links the judicial exception to a field of use and/or technology environment, see MPEP 2106.05(h)).
utilizing an asynchronous messaging service to communicate model states and events
(i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation directed to mere data gathering as deemed insufficient to transform the judicial exception because claimed elements are considered insignificant extra-solution activity, See MPEP (2106.05(g))).
wherein the respective stop criteria includes a prediction duration threshold and is based on one or more central processing unit restrictions associated with each neural network in the plurality of neural networks;
(i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation which simply links the judicial exception to a field of use and/or technology environment, see MPEP 2106.05(h)).
wherein the aggregated prediction only includes predictions that do not exceed the prediction duration threshold;
(i.e.: deemed insufficient to transform the judicial exception to a patentable invention because the claim recites limitation which simply links the judicial exception to a field of use and/or technology environment, see MPEP 2106.05(h)).
In Step 2B: This part of the eligibility analysis evaluates whether the claim as a whole amount to significantly more than the recited exception, i.e., whether any additional element, or combination of additional elements, adds an inventive concept to the claim. MPEP 2106.05.
First, the additional limitations (III, IV, V, VII and VIII) and the preamble are deemed insufficient to transform the judicial exception to a patentable invention to a patentable invention because they generally link the judicial exception to the technology environment, see MPEP 2106.05(h).
Second, the additional limitations (I) recite mere application of the abstract idea or mere instructions to implement an abstract idea on a computer are deemed insufficient to transform the judicial exception to a patentable invention to a patentable invention because the limitations generally apply the use of a generic computer and/or process with the judicial exception, see MPEP 2106.05(f).
Third, additional limitation (II and VI) recites extra/post solution activity, as analyzed above, are activity that are well-understood routine and conventional, specifically: the courts have recognized the computer functions as well‐understood, routine, and conventional functions.
Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TL| Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network). See MPEP 2106.05(d)(II).
As analyzed above, the additional elements, analyzed above, do not integrate the noted judicial exception into a practical application because they do not impose any meaningful limits on practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Regarding claim 8,
A computer program product comprising: one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the stored program instructions comprising
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and are considered to adding the words “apply it” (or an equivalent) with the judicial exception, See MPEP 2106.05(f).
The rest of the limitations are analogous to claim 1, so are rejected under similar rationale.
Regarding claim 15
A computer system comprising: one or more computer processors; one or more computer readable storage media; and program instructions stored on the computer readable storage media for execution by at least one of the one or more processors, the stored program instructions comprising
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and are considered to adding the words “apply it” (or an equivalent) with the judicial exception, See MPEP 2106.05(f).
The rest of the limitations are analogous to claim 1, so are rejected under similar rationale.
Regarding claim 2, dependent on claim 1, and fail to resolve the deficiencies identified above by
integrating the judicial exception into a practical application, or introducing significantly more than
the judicial exception. The claim recites:
generating, by one or more computer processors, the aggregated prediction utilizing each neural network in the determined plurality of neural networks subject to adjusted stop criteria
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to a mental process of evaluating and judgment of determining stopping criteria of a model training.
Regarding claims 9 and 16 recite similar limitations as claim 2, so is rejected under the same rationale.
Regarding claim 3, depends on claim 1, and fail to resolve the deficiencies identified above by integrating the judicial exception into a practical application, or introducing significantly more than the judicial exception. The claim recites:
responsive to the calculated confidence value for the aggregated prediction reaching a confidence threshold, deploying, by one or more computer processors, the plurality of neural networks,
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and are considered to adding the words “apply it” (or an equivalent) with the judicial exception, See MPEP 2106.05(f).
Limitations directed to using the computer as a tool for implementing an abstract idea cannot integrate a judicial exception into a practical application at Step 2A or provide an inventive concept in Step 2B.
Claim(s) 10 and 17 recite similar limitations as claim 3, so is rejected under the same rationale.
Regarding claim 4, depends on claim 3, and fail to resolve the deficiencies identified above by integrating the judicial exception into a practical application, or introducing significantly more than the judicial exception. The claim recites:
labeling, by one or more computer processors, one or more unlabeled datapoints with the deployed plurality of neural networks,
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to a mental process of evaluating and judgment of determining unlabeled dataset.
Claim(s) 11 and 18 recite similar limitations as claim 4, so is rejected under the same rationale.
Regarding claim 5, depends on claim 1, and fail to resolve the deficiencies identified above by integrating the judicial exception into a practical application, or introducing significantly more than the judicial exception. The claim recites:
training, by one or more computer processors, the plurality of neural networks utilizing the determined features and associated training data,
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and are considered to adding the words “apply it” (or an equivalent) with the judicial exception, See MPEP 2106.05(f).
Limitations directed to using the computer as a tool for implementing an abstract idea cannot integrate a judicial exception into a practical application at Step 2A or provide an inventive concept in Step 2B.
Claim(s) 12 and 19 recite similar limitations as claim 5, so is rejected under the same rationale.
Regarding claim 6, depends on claim 2, and fail to resolve the deficiencies identified above by integrating the judicial exception into a practical application, or introducing significantly more than the judicial exception. The claim recites:
clustering, by one or more computer processors, the plurality of neural networks; and
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to a mental process of evaluating and judgment of combining predictions from multiple models.
identifying, by one or more computer processors, one or more neural networks with high confidence predictions utilizing the clustered plurality of models,
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to a mental process of making a judgement on which model has the highest confidence prediction based on combined prediction.
Claim(s) 13 and 20 recite similar limitations as claim 6, so is rejected under the same rationale.
Regarding claim 7, depends on claim 1, and fail to resolve the deficiencies identified above by integrating the judicial exception into a practical application, or introducing significantly more than the judicial exception, the claim recites:
monitoring,
Deemed insufficient to transform the judicial exception to a patentable invention because the limitation is directed to a mental process of making observation.
by one or more computer processors, one or more neural networks utilizing a publish and subscribe structure
The recitation in the additional limitation directed to mere data gathering as deemed insufficient to transform the judicial exception because claimed elements are considered insignificant extra-solution activity and well-understood routine and conventional (2106.05(d)).
Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TL| Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) (computer receives and sends information over a network). See MPEP 2106.05(d)(II).
The additional limitations as analyze failed to integrate a judicial exception into a practical application at Step 2A and provide an inventive concept in Step 2B, per the analysis above.
Claim 14 recites similar limitations as claim 7, so is rejected under the same rationale.
Allowable Subject Matter
Claim(s) 1 – 20 would be allowable if rewritten or amended to overcome the rejections under 35 U.S.C. 101 and 35 U.S.C. 112 set forth in this Office action. The following is a statement of reasons for the indication of allowable subject matter:
Claim(s) 1, 8 and 15 as a whole with regards to technical features recited by the claim limitations including directed to:
A computer-implemented method comprising: responsive to streaming data, continuously initiating, by one or more computer processors, a plurality of parallel-operating machine learning pipelines with the streaming data,
wherein the plurality of parallel-operating machine learning pipelines comprises a plurality of neural networks, wherein each neural network in the plurality of neural networks has a distinctive stop criteria specific to a neural network structure respectively associated with each neural network in the plurality of neural networks,
wherein weights of a previously trained model that failed to exceed a confidence threshold are utilized;
determining, by one or more computer processors, one or more neural networks within the plurality of neural networks to incorporate a plurality of determined features from the streaming data, comprising: calculating, by one or more computer processors, a reduction in entropy from a transformation of the dataset based on an information gain of each feature in the plurality of determined features;
generating, by one or more computer processors, an aggregated prediction utilizing each neural network in the plurality of neural networks, in parallel, in the determined plurality of neural networks subject to the respective stop criteria utilizing an asynchronous messaging service to communicate model states and events,
wherein the respective stop criteria includes a prediction duration threshold and is based on one or more central processing unit restrictions associated with each neural network in the plurality of neural networks, wherein the aggregated prediction only includes predictions that do not exceed the prediction duration threshold;
calculating, by one or more computer processors, a confidence value for the aggregated prediction; and
responsive to the calculated confidence value for the aggregated prediction not reaching a confidence threshold, adjusting, by one or more computer processors, the respective stop criteria to increase a prediction duration to allow for computationally intensive neural networks to finish predictions to contribute to the aggregated prediction.
Closest prior arts disclose:
Nair, Pub. No.: US20200372342A1, (2019-12-09).
Nair outlines a process for training multiple neural networks with different hyperparameters, where each training run is monitored and may be stopped early if further improvement is unlikely. A probability threshold and patience value are used to decide when to halt training, optimizing resource use by focusing on promising models. However, Nair does not teach incorporating stop criteria, utilizing asynchronous messaging service to communicate model states and events and allowing partial results from incomplete pipelines to contribute to final predictions, reduces computational time and resource usage while outputting predictions that do not exceed a predetermined threshold. For processes that exceeds the threshold, adjust the stop criteria to allow the duration for computationally intensive neural networks to finish prediction to contribute to the aggregated output prediction.
MUELLER, Pub.: No.: US20210326717A1, (2020-04-15).
A system that monitors machine learning model training by generating and storing performance metrics. User can access these metrics in rea time to decide whether to adjust or stop training, such as when an error rate is high, confidence is low or performance declines. Based on user input, the system can modify or restart the training process by updating the model’s container, algorithm or parameter or stop training algorithm by deleting the model and associated data. However, MUELLER does not teach incorporating stop criteria, utilizing asynchronous messaging service to communicate model states and events and allowing partial results from incomplete pipelines to contribute to final predictions, reduces computational time and resource usage while outputting predictions that do not exceed a predetermined threshold. For processes that exceeds the threshold, adjust the stop criteria to allow the duration for computationally intensive neural networks to finish prediction to contribute to the aggregated output prediction.
Teerapittayanon, et al., "Branchynet: Fast inference via early exiting from deep neural networks.", (2016).
This paper highlights three main contributions. First, it introduces Fast Inference with Early Exit Branches, where BranchyNet allows many samples to exit early, reducing computation and I/O costs, leading to faster runtime and lower energy use. Second, it employs Regularization via Joint Optimization, where BranchyNet optimizes the weighted loss at all exit points, helping to prevent overfitting and enhance test accuracy. Lastly, it addresses the Mitigation of Vanishing Gradients, as early exit points provide stronger gradient signals during backpropagation, which results in better features and improved accuracy in lower layers. However, Teerapittayanon does not teach incorporating stop criteria, utilizing asynchronous messaging service to communicate model states and events and allowing partial results from incomplete pipelines to contribute to final predictions, reduces computational time and resource usage while outputting predictions that do not exceed a predetermined threshold. For processes that exceeds the threshold, adjust the stop criteria to allow the duration for computationally intensive neural networks to finish prediction to contribute to the aggregated output prediction.
Conclusion
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 MATIYAS T MARU whose telephone number is (571)270-0902. The examiner
can normally be reached Monday - Friday (8:00am - 4:00pm) 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,
Michelle Bechtold can be reached on (571)431-0762. The fax phone number for the organization were 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.
/M.T.M./ Examiner, Art Unit 2148
/MICHELLE T BECHTOLD/ Supervisory Patent Examiner, Art Unit 2148
Prosecution Insights