Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The Amendment filed 11/19/2025 has been entered. Claims 1-20 remain pending in this application.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 4, 6-10, 12, 14-17 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over SHALOUDEGI et al. (US 20230117768 A1 hereinafter SHALOUDEGI) in view of Rawashdeh et al. (US 20230177404 A1 hereinafter Rawashdeh) and Balakrishnan et al. (US 20230177349 A1 hereinafter Balakrishnan)
As to independent claim 1, SHALOUDEGI teaches a server comprising: [server ¶47]
a controller programmed to: [processing unit on system/server ¶49-50]
obtain local gradients received from a plurality of vehicles [federated learning with autonomous vehicles ¶147, local model gradients ¶59 ". Information about the updated parameters of the local model are sent back to the central server 110 by the user devices 102, typically in the form of gradients"] related to traffic, construction, or pedestrian behavior, to promote safe driving)"]
determine weights for local gradients received from the plurality of vehicles based on the determined contributions; [weighted proximal maps ¶149 and gradients ¶59 "Information about the updated parameters of the local model are sent back to the central server 110 by the user devices 102, typically in the form of gradients"]
aggregate the local gradients based on the adjusted weights to obtain a global model. [aggregates the information (gradients) for a global model ¶59 " The central server 110 aggregates the received information to update the parameters of the global model. In the case of FedAvg, the update is performed by averaging the received gradients and adding the average of the received gradients to the current model parameters."]
SHALOUDEGI does not specifically teach adjust the weights based on a comparison of potential functions for the plurality of vehicles and transmit the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model.
However, Rawashdeh teaches adjust the weights based on a comparison of potential functions for the plurality of vehicles; [looks at the loss functions (potential) from clients (compares loss from vehicle functions) ¶55-57 "examines the loss for each client at 305 and determines the loss reduction (change in loss over the number of cycles, for example) at 307. The process also determines whether the client qualifies for inclusion in training at 309, which in this case is a comparison of the loss reduction to whatever value is set for gamma."]
transmit the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model. [sends global model to vehicles ¶33-34 "In federated learning, globally trained models may be validated in the cloud prior to distribution or may be distributed to a limited subset of vehicles for testing/validation prior to wide distribution. The in-vehicle training process allows for vehicles to self-improve their own models using data that is relevant to both the model and to the given vehicle. Because vehicles will observe significantly varied information, based on localities, overall usage, specific types of usage, owner/operator tendencies, environmental variances, etc"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimized algorithm disclosed by SHALOUDEGI by incorporating the adjust the weights based on a comparison of potential functions for the plurality of vehicles and transmit the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model disclosed by Rawashdeh because both techniques address the same field of machine learning and by incorporating Rawashdeh into SHALOUDEGI further improves the centralized or global model to be more accurate and optimized [Rawashdeh ¶19]
SHALOUDEGI and Rawashdeh do not specifically teach determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients;
However, Balakrishnan teaches determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients;
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI and Rawashdeh by incorporating the determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients disclosed by Balakrishnan because all techniques address the same field of machine learning and by incorporating Balakrishnan into SHALOUDEGI and Rawashdeh improves global models for accuracy offering better orchestration and management of applications [Balakrishnan ¶4-5]
As to dependent claim 2, the rejection of claim 1 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach calculate a utility function based on the contributions and the weights; and [SHALOUDEGI optimization algorithm that minimizing the expected value of a weighted sum of the loss functions of all the user devices, by considering the contributions (loss functions) and their weights for a new model ¶64-65]
calculate the potential function for each of the plurality of vehicles based on the utility function. [Rawashdeh calculates global model (has a loss function) is distributed to clients/vehicles (for each) ¶57, ¶26 "distribute the updated global model to vehicles "]
As to dependent claim 4, the rejection of claim 1 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach increase the weight for a first vehicle and decrease the weight for a second vehicle in response to determining that a value of the potential function for the first vehicle is greater than a value of the potential function for the second vehicle. [Rawashdeh compares loss of different vehicles and picks better or faster ones (greater value) ¶28 "compare loss reduction to its own loss reduction over a comparable span, and once sufficient shared models representing “better” loss reduction have been received, the given vehicle can train its own model using federated averaging of the weights associated with the faster-learning models"]
As to dependent claim 6, the rejection of claim 1 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach obtain a plurality of local gradients from the plurality of vehicles; and [SHALOUDEGI gradients sent to server ¶59]
determine the contributions of the plurality of vehicles based on differences between the local gradients and a global gradient vector. [Balakrishnan cosine similarity (angle based score) of the gradients between local and global ¶224]
As to dependent claim 7, the rejection of claim 1 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach calculate a utility function based on the contributions and the weights; and [SHALOUDEGI optimization algorithm that minimizing the expected value of a weighted sum of the loss functions of all the user devices, by considering the contributions (loss functions) and their weights for a new model ¶64-65]
repeat adjusting the weights until the utility function is maximized. [SHALOUDEGI repeated ¶127, iterations/rounds ¶59-¶61]
As to dependent claim 8, the rejection of claim 1 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach transmit the global model to the plurality of vehicles. [Rawashdeh ¶26 "distribute the updated global model to vehicles 101, 121, 141 as appropriate"]
As to independent claim 9, SHALOUDEGI teaches a method for aggregating models from a plurality of vehicles, the method comprising: [vehicles that help build a global model ¶147-149]
obtaining local gradients received from a plurality of vehicles [federated learning with autonomous vehicles ¶147, local model gradients ¶59 ". Information about the updated parameters of the local model are sent back to the central server 110 by the user devices 102, typically in the form of gradients"] related to traffic, construction, or pedestrian behavior, to promote safe driving)"]
determining weights for local gradients received from the plurality of vehicles based on the determined contributions; [weighted proximal maps ¶149 and gradients ¶59 "Information about the updated parameters of the local model are sent back to the central server 110 by the user devices 102, typically in the form of gradients"]
aggregating the local gradients based on the adjusted weights to obtain a global model. [aggregates the information (gradients) for a global model ¶59 " The central server 110 aggregates the received information to update the parameters of the global model. In the case of FedAvg, the update is performed by averaging the received gradients and adding the average of the received gradients to the current model parameters."]
SHALOUDEGI does not specifically teach adjusting the weights based on a comparison of potential functions for the plurality of vehicles and transmitting the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model.
However, Rawashdeh teaches adjusting the weights based on a comparison of potential functions for the plurality of vehicles; [looks at the loss functions (potential) from clients (compares loss from vehicle functions) ¶55-57 "examines the loss for each client at 305 and determines the loss reduction (change in loss over the number of cycles, for example) at 307. The process also determines whether the client qualifies for inclusion in training at 309, which in this case is a comparison of the loss reduction to whatever value is set for gamma."]
transmitting the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model. [sends global model to vehicles ¶33-34 "In federated learning, globally trained models may be validated in the cloud prior to distribution or may be distributed to a limited subset of vehicles for testing/validation prior to wide distribution. The in-vehicle training process allows for vehicles to self-improve their own models using data that is relevant to both the model and to the given vehicle. Because vehicles will observe significantly varied information, based on localities, overall usage, specific types of usage, owner/operator tendencies, environmental variances, etc"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimized algorithm disclosed by SHALOUDEGI by incorporating the adjust the weights based on a comparison of potential functions for the plurality of vehicles and transmit the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model disclosed by Rawashdeh because both techniques address the same field of machine learning and by incorporating Rawashdeh into SHALOUDEGI further improves the centralized or global model to be more accurate and optimized [Rawashdeh ¶19]
SHALOUDEGI and Rawashdeh do not specifically teach determining contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients;
However, Balakrishnan teaches determining contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients;
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI and Rawashdeh by incorporating the determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients disclosed by Balakrishnan because all techniques address the same field of machine learning and by incorporating Balakrishnan into SHALOUDEGI and Rawashdeh improves global models for accuracy offering better orchestration and management of applications [Balakrishnan ¶4-5]
As to dependent claim 10, the rejection of claim 9 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach calculating a utility function based on the contributions and the weights; and [SHALOUDEGI optimization algorithm that minimizing the expected value of a weighted sum of the loss functions of all the user devices, by considering the contributions (loss functions) and their weights for a new model ¶64-65]
calculating the potential function for each of the plurality of vehicles based on the utility function. [Rawashdeh calculates global model (has a loss function) is distributed to clients/vehicles (for each) ¶57, ¶26 "distribute the updated global model to vehicles "]
As to dependent claim 12, the rejection of claim 9 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach increasing the weight for a first vehicle and decrease the weight for a second vehicle in response to determining that a value of the potential function for the first vehicle is greater than a value of the potential function for the second vehicle. [Rawashdeh compares loss of different vehicles and picks better or faster ones (greater value) ¶28 "compare loss reduction to its own loss reduction over a comparable span, and once sufficient shared models representing “better” loss reduction have been received, the given vehicle can train its own model using federated averaging of the weights associated with the faster-learning models"]
As to dependent claim 14, the rejection of claim 9 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach obtain a plurality of local gradients from the plurality of vehicles; and [SHALOUDEGI gradients sent to server ¶59]
determine the contributions of the plurality of vehicles based on differences between the local gradients and a global gradient vector. [Balakrishnan cosine similarity (angle based score) of the gradients between local and global ¶224]
As to dependent claim 15, the rejection of claim 9 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach calculate a utility function based on the contributions and the weights; and [SHALOUDEGI optimization algorithm that minimizing the expected value of a weighted sum of the loss functions of all the user devices, by considering the contributions (loss functions) and their weights for a new model ¶64-65]
repeat adjusting the weights until the utility function is maximized. [SHALOUDEGI repeated ¶127, iterations/rounds ¶59-¶61]
As to independent claim 16, SHALOUDEGI teaches a system comprising: [server system ¶47-48]
a plurality of vehicles; and [vehicles as user devices ¶43]
a server comprising a controller programmed to: [processing unit on system/server ¶49-50]
obtain local gradients received from a plurality of vehicles [federated learning with autonomous vehicles ¶147, local model gradients ¶59 ". Information about the updated parameters of the local model are sent back to the central server 110 by the user devices 102, typically in the form of gradients"] related to traffic, construction, or pedestrian behavior, to promote safe driving)"]
determine weights for local gradients received from the plurality of vehicles based on the determined contributions; [weighted proximal maps ¶149 and gradients ¶59 "Information about the updated parameters of the local model are sent back to the central server 110 by the user devices 102, typically in the form of gradients"]
aggregate the local gradients based on the adjusted weights to obtain a global model. [aggregates the information (gradients) for a global model ¶59 " The central server 110 aggregates the received information to update the parameters of the global model. In the case of FedAvg, the update is performed by averaging the received gradients and adding the average of the received gradients to the current model parameters."]
SHALOUDEGI does not specifically teach adjust the weights based on a comparison of potential functions for the plurality of vehicles and transmit the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model.
However, Rawashdeh teaches adjust the weights based on a comparison of potential functions for the plurality of vehicles; [looks at the loss functions (potential) from clients (compares loss from vehicle functions) ¶55-57 "examines the loss for each client at 305 and determines the loss reduction (change in loss over the number of cycles, for example) at 307. The process also determines whether the client qualifies for inclusion in training at 309, which in this case is a comparison of the loss reduction to whatever value is set for gamma."]
transmit the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model. [sends global model to vehicles ¶33-34 "In federated learning, globally trained models may be validated in the cloud prior to distribution or may be distributed to a limited subset of vehicles for testing/validation prior to wide distribution. The in-vehicle training process allows for vehicles to self-improve their own models using data that is relevant to both the model and to the given vehicle. Because vehicles will observe significantly varied information, based on localities, overall usage, specific types of usage, owner/operator tendencies, environmental variances, etc"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimized algorithm disclosed by SHALOUDEGI by incorporating the adjust the weights based on a comparison of potential functions for the plurality of vehicles and transmit the global model to the plurality of vehicles and instruct the plurality of vehicles to autonomously operate based on the global model disclosed by Rawashdeh because both techniques address the same field of machine learning and by incorporating Rawashdeh into SHALOUDEGI further improves the centralized or global model to be more accurate and optimized [Rawashdeh ¶19]
SHALOUDEGI and Rawashdeh do not specifically teach determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients;
However, Balakrishnan teaches determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients;
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI and Rawashdeh by incorporating the determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients disclosed by Balakrishnan because all techniques address the same field of machine learning and by incorporating Balakrishnan into SHALOUDEGI and Rawashdeh improves global models for accuracy offering better orchestration and management of applications [Balakrishnan ¶4-5]
As to dependent claim 17, the rejection of claim 16 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach calculate a utility function based on the contributions and the weights; and [SHALOUDEGI optimization algorithm that minimizing the expected value of a weighted sum of the loss functions of all the user devices, by considering the contributions (loss functions) and their weights for a new model ¶64-65]
calculate the potential function for each of the plurality of vehicles based on the utility function. [Rawashdeh calculates global model (has a loss function) is distributed to clients/vehicles (for each) ¶57, ¶26 "distribute the updated global model to vehicles "]
As to dependent claim 19, the rejection of claim 16 is incorporated. SHALOUDEGI, Rawashdeh and Balakrishnan further teach increase the weight for a first vehicle and decrease the weight for a second vehicle in response to determining that a value of the potential function for the first vehicle is greater than a value of the potential function for the second vehicle. [Rawashdeh compares loss of different vehicles and picks better or faster ones (greater value) ¶28 "compare loss reduction to its own loss reduction over a comparable span, and once sufficient shared models representing “better” loss reduction have been received, the given vehicle can train its own model using federated averaging of the weights associated with the faster-learning models"]
Claims 3, 11 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over SHALOUDEGI in view of Rawashdeh and Balakrishnan, as applied in claim 2, 10 and 17 above, and further in view of JUNG et al. (US 20250119245 A1 hereinafter Jung)
As to dependent claim 3, SHALOUDEGI, Rawashdeh and Balakrishnan teach the rejection of claim 2 is incorporated.
SHALOUDEGI, Rawashdeh and Balakrishnan further teach wherein the utility function comprises an average contribution function of the weights and the contributions and [SHALOUDEGI averaging ¶4, ¶59]
SHALOUDEGI, Rawashdeh and Balakrishnan do not specifically teach a Shannon entropy function of the weights.
However, Jung teaches a Shannon entropy function of the weights. [Shannon entropy ¶128 "Referring to FIG. 9, the Shannon entropy H(W) of the world model W may be represented as follows"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI, Rawashdeh and Balakrishnan by incorporating the Shannon entropy function of the weights disclosed by Jung because all techniques address the same field of machine learning and by incorporating Jung into SHALOUDEGI, Rawashdeh and Balakrishnan protects the privacy of data in machine learning [Jung ¶83]
As to dependent claim 11, SHALOUDEGI, Rawashdeh and Balakrishnan teach the rejection of claim 10 is incorporated.
SHALOUDEGI, Rawashdeh and Balakrishnan further teach wherein the utility function comprises an average contribution function of the weights and the contributions and [SHALOUDEGI averaging ¶4, ¶59]
SHALOUDEGI, Rawashdeh and Balakrishnan do not specifically teach a Shannon entropy function of the weights.
However, Jung teaches a Shannon entropy function of the weights. [Shannon entropy ¶128 "Referring to FIG. 9, the Shannon entropy H(W) of the world model W may be represented as follows"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI, Rawashdeh and Balakrishnan by incorporating the Shannon entropy function of the weights disclosed by Jung because all techniques address the same field of machine learning and by incorporating Jung into SHALOUDEGI, Rawashdeh and Balakrishnan protects the privacy of data in machine learning [Jung ¶83]
As to dependent claim 18, SHALOUDEGI, Rawashdeh and Balakrishnan teach the rejection of claim 17 is incorporated.
SHALOUDEGI, Rawashdeh and Balakrishnan further teach wherein the utility function comprises an average contribution function of the weights and the contributions and [SHALOUDEGI averaging ¶4, ¶59]
SHALOUDEGI, Rawashdeh and Balakrishnan do not specifically teach a Shannon entropy function of the weights.
However, Jung teaches a Shannon entropy function of the weights. [Shannon entropy ¶128 "Referring to FIG. 9, the Shannon entropy H(W) of the world model W may be represented as follows"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI, Rawashdeh and Balakrishnan by incorporating the Shannon entropy function of the weights disclosed by Jung because all techniques address the same field of machine learning and by incorporating Jung into SHALOUDEGI, Rawashdeh and Balakrishnan protects the privacy of data in machine learning [Jung ¶83]
Claims 5, 13 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over SHALOUDEGI in view of Rawashdeh and Balakrishnan, as applied in claim 4, 12 and 19 above, and further in view of Acharya et al. (US 20210295215 A1 hereinafter Acharya).
As to dependent claim 5, SHALOUDEGI, Rawashdeh and Balakrishnan teach the rejection of claim 4 is incorporated.
SHALOUDEGI, Rawashdeh and Balakrishnan do not specifically teach wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle.
However, Acharya teaches wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle. [weight is a function of size/quality¶48 "weight for a given model may be a function of an accuracy of the model on the data stream, the size of the data stream, and an expected quality of the data stream"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI, Rawashdeh and Balakrishnan by incorporating the wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle disclosed by Acharya because all techniques address the same field of machine learning and by incorporating Acharya into SHALOUDEGI, Rawashdeh and Balakrishnan helps reduce costs and traffic while keeping privacy [Acharya ¶3, ¶22]
As to dependent claim 13, SHALOUDEGI, Rawashdeh and Balakrishnan teach the rejection of claim 12 is incorporated.
SHALOUDEGI, Rawashdeh and Balakrishnan do not specifically teach wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle.
However, Acharya teaches wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle. [weight is a function of size/quality¶48 "weight for a given model may be a function of an accuracy of the model on the data stream, the size of the data stream, and an expected quality of the data stream"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI, Rawashdeh and Balakrishnan by incorporating the wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle disclosed by Acharya because all techniques address the same field of machine learning and by incorporating Acharya into SHALOUDEGI, Rawashdeh and Balakrishnan helps reduce costs and traffic while keeping privacy [Acharya ¶3, ¶22]
As to dependent claim 20, SHALOUDEGI, Rawashdeh and Balakrishnan teach the rejection of claim 19 is incorporated.
SHALOUDEGI, Rawashdeh and Balakrishnan do not specifically teach wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle.
However, Acharya teaches wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle. [weight is a function of size/quality¶48 "weight for a given model may be a function of an accuracy of the model on the data stream, the size of the data stream, and an expected quality of the data stream"]
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify the optimization algorithm disclosed by SHALOUDEGI, Rawashdeh and Balakrishnan by incorporating the wherein an amount of the increase of the weight is proportional to a difference between the value of the potential function for the first vehicle and the value of the potential function for the second vehicle disclosed by Acharya because all techniques address the same field of machine learning and by incorporating Acharya into SHALOUDEGI, Rawashdeh and Balakrishnan helps reduce costs and traffic while keeping privacy [Acharya ¶3, ¶22]
Response to Arguments
Applicant's arguments filed 11/19/2025 with respect to 101, those rejections have been withdrawn.
Applicant's arguments filed 11/19/2025. In the remark, applicant argues that:
(1) SHALOUDEGI, Rawashdeh and Zhou fail to teach “determine contributions of the plurality of vehicles in a federated learning framework based on angles between a global gradient and the local gradients;” as recited by amended claim 1, 9 and 16.
As to point (1), Applicant’s arguments with respect to claim 1 have been considered but are moot in view of a new ground of rejection as set forth above of SHALOUDEGI in view of Rawashdeh and Balakrishnan.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 C.F.R. § 1.111(c) to consider these references fully when responding to this action.
Zhou et al. (US 12175338 B2) teaches aggregating gradients from workers and calculating weights (see Col. 2 ln. 23-35).
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BEAU SPRATT whose telephone number is (571)272-9919. The examiner can normally be reached M-F 8:30-5 PST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jennifer Welch can be reached on 5712127212. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/BEAU D SPRATT/Primary Examiner, Art Unit 2143
Prosecution Insights