DETAILED ACTION
This action is responsive to the application filed on 03/06/2024. Claims 1-20 are pending in the case. Claims 1, 14, and 18 are independent claims.
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 .
Priority
Acknowledgement is made of applicant’s claim for domestic priority based on a provisional application filed on 05/31/2023.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 01/16/2025 is being considered by the examiner.
Specification
The abstract of the disclosure is objected to because of. The abstract exceeds the 150-word maximum. A corrected abstract of the disclosure is required and must be presented on a separate sheet, apart from any other text. See MPEP § 608.01(b).
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: "an incentive calculator communicably coupled to the group manager to calculate..." in claim 1.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
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.
Claims 1-13 and 18-20 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 applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim limitation “an incentive calculator communicably coupled to the group manager to calculate...” invokes 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed function and to clearly link the structure, material, or acts to the function. Applicant’s specification paragraphs 0070-0072 discloses a generic computer. Applicant’s specification paragraphs 0047-0048, 0052-0054, 0058, 0060 include examples of incentive calculations, however, there is no detailed algorithm for performing the calculation. Thus, no algorithm is disclosed by the applicant to provide corresponding acts to the function. Therefore, the claim is indefinite and is rejected under 35 U.S.C. 112(b) or pre-AIA 35 U.S.C. 112, second paragraph.
Applicant may:
(a) Amend the claim so that the claim limitation will no longer be interpreted as a limitation under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph;
(b) Amend the written description of the specification such that it expressly recites what structure, material, or acts perform the entire claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(c) Amend the written description of the specification such that it clearly links the structure, material, or acts disclosed therein to the function recited in the claim, without introducing any new matter (35 U.S.C. 132(a)).
If applicant is of the opinion that the written description of the specification already implicitly or inherently discloses the corresponding structure, material, or acts and clearly links them to the function so that one of ordinary skill in the art would recognize what structure, material, or acts perform the claimed function, applicant should clarify the record by either:
(a) Amending the written description of the specification such that it expressly recites the corresponding structure, material, or acts for performing the claimed function and clearly links or associates the structure, material, or acts to the claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(b) Stating on the record what the corresponding structure, material, or acts, which are implicitly or inherently set forth in the written description of the specification, perform the claimed function. For more information, see 37 CFR 1.75(d) and MPEP §§ 608.01(o) and 2181.
Claims 2-13 are rejected as being dependent upon a rejected base claim without curing any of the deficiencies.
Claim 3 recites the limitation “the model” in line 1. There is insufficient antecedent basis for this limitation in the claim. Claim 1 recites “a global model” and claim 2 recites “a model format”. It is unclear if applicant is attempting to recite a new claim element or if applicant is attempting to refer to a previously recited claim element. For examination purposes, this limitation is being interpreted as “the model format”, referring to the previously recited limitation of claim 2.
Claim 12 recites the limitation "the inferencing service" in line 1. There is insufficient antecedent basis for this limitation in the claim. It is unclear if applicant is attempting to refer to a previously recited claim element or if applicant is attempting to recite a new claim element. For examination purposes, this limitation is being interpreted to mean “an inferencing service” reciting a new claim element.
Claim 13 recites the limitation "the inferencing service" in line 1. There is insufficient antecedent basis for this limitation in the claim. It is unclear if applicant is attempting to refer to a previously recited claim element or if applicant is attempting to recite a new claim element. For examination purposes, this limitation is being interpreted to mean “an inferencing service” reciting a new claim element.
Claim 18 recites the limitation "the " in line3-4. There is insufficient antecedent basis for this limitation in the claim. The claim also recites “perform operations for building a global machine learning (ML) model”. It is unclear if applicant is attempting to refer to a previously recited claim element or if applicant is attempting to recite a new claim element. For examination purposes, this limitation is being interpreted to mean “the operations” referring to a previously recited claim element.
Claims 19-20 are rejected as being dependent upon a rejected base claim without curing any of the deficiencies.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding claim 1:
Step 1 Statutory Category: Claim 1 is directed to a machine which falls under one of the four statutory categories.
Step 2A Prong 1 Judicial exception: Claim 1 recites, in part, “identify a group of clients to build a global model”. This limitation, under the broadest reasonable interpretation, covers the recitation of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion), in this case an observation. See MPEP § 2106.04(a)(2)(III). Further, the claim recites: “build the global model by … aggregating model parameters received from the group to build the global model, the model parameters being generated by the clients training a local ML model at their respective client sites using local data at their respective client sites”. This limitation, under the broadest reasonable interpretation, covers the recitation of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. A mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number”. See MPEP §2106.04(a)(2)(I)(C). Further, the claim recites: “calculate an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data”. This limitation, under the broadest reasonable interpretation, and in light of applicant’s specification paragraphs 0047-0048, covers the recitation of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. A mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number”. See MPEP §2106.04(a)(2)(I)(C).
Step 2A Prong 2 Integration into a practical application: This judicial exception is not integrated into a practical application. In particular the claim recites: “a global machine learning (ML) model” and “by federated learning”. These limitations are additional elements that amount to generally linking the use of the judicial exception to a particular technological environment or field of use. See MPEP §2106.05(h). Further, the claim recites: “an architecture”, “a platform”, “a group manager”, and “an incentive calculator communicably coupled to the group manager”. These limitations are additional elements that amount to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §2106.05(f). Further, the claim recites: “supplying a model definition for the global model to the group”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g).
Step 2B Significantly more: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements: “a global machine learning (ML) model” and “by federated learning” amount to generally linking the use of the judicial exception to a particular technological environment or field of use. Elements that merely amount to generally linking the use of the judicial exception to a particular technological environment or field of use cannot provide an inventive concept. Further, the additional elements: “an architecture”, “a platform”, “a group manager”, and “an incentive calculator communicably coupled to the group manager” amount to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. Elements that merely amount to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process cannot provide an inventive concept. Further, the additional element “supplying a model definition for the global model to the group” amounts adding insignificant extra-solution activity to the judicial exception, and further, is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 2, the rejection of claim 1 is incorporated, and further, the claim recites: “registries for storing a model format and feature format for the global model” and “a server to provide the model format and feature format to the group as part of the model definition”. These limitations are additional elements that amount to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §2106.05(f). Elements that merely amount to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process cannot provide an inventive concept. The claim is not patent eligible.
Regarding claim 3, the rejection of claim 2 is incorporated, and further, the claim recites: “wherein the server provides the model and feature format to the group as part of the model definition after the group manager sends a request for training to the clients and the clients join the group in response to the request for training”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 4, the rejection of claim 2 is incorporated, and further, the claim recites: “wherein the server provides the model format and feature format to said each client for use when training their local model”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 5, the rejection of claim 1 is incorporated, and further, the claim recites: “wherein the group manager is operable to aggregate the model parameters from the clients based on at least one of client data set size of data used to train their respective local model and their local model’s accuracy”. This limitation is a continuation of the “build the global model by … aggregating model parameters received from the group to build the global model, the model parameters being generated by the clients training a local ML model at their respective client sites using local data at their respective client sites” limitation identified as an abstract idea in the rejection of the parent claim. Thus, the claim recites a judicial exception.
The claim does not include any additional elements that amount to an integration of the judicial exception into a practical application, nor to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 6, the rejection of claim 1 is incorporated, and further, the claim recites: “wherein the incentive calculator is operable to calculate the incentive based on an amount of each client’s contribution”. This limitation is a continuation of the “calculate an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data” limitation identified in the rejection of the parent claim. Thus, the claim recites a judicial exception.
The claim does not include any additional elements that amount to an integration of the judicial exception into a practical application, nor to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 7, the rejection of claim 6 is incorporated, and further, the claim recites: “wherein each client’s contribution is based on at least one of client data set size of data used to train their respective local model and their local model’s accuracy”. This limitation is a continuation of the “wherein the incentive calculator is operable to calculate the incentive based on an amount of each client’s contribution” limitation identified in the rejection of the parent claim. Thus, the claim recites a judicial exception.
The claim does not include any additional elements that amount to an integration of the judicial exception into a practical application, nor to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 8, the rejection of claim 1 is incorporated, and further, the claim recites: “wherein the incentive calculator calculates the incentive based on stored training and trading histories”. This limitation is a continuation of the “calculate an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data” limitation identified in the rejection of the parent claim. Thus, the claim recites a judicial exception.
The claim does not include any additional elements that amount to an integration of the judicial exception into a practical application, nor to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 9, the rejection of claim 1 is incorporated, and further, the claim recites: “wherein the local model trained by the clients has identical feature sets”. This limitation is an additional element that amounts to generally linking the use of the judicial exception to a particular technological environment or field of use. See MPEP §2106.05(h). Elements that merely generally link the use of the judicial exception to a particular technological environment or field of use cannot provide an inventive concept. The claim is not patent eligible.
Regarding claim 10, the rejection of claim 1 is incorporated, and further, the claim recites: “wherein the platform is operable to identify the group to build a global model in response to a request from one or more users of the global model”. This limitation is a continuation of the “identify a group of clients to build a global model” limitation identified as an abstract idea in the rejection of the parent claim. Thus, the claim recites a judicial exception.
The claim does not include any additional elements that amount to an integration of the judicial exception into a practical application, nor to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 11, the rejection of claim 1 is incorporated, and further, the claim recites: “an inference service responsive to request for use of the global model by one or more users”. This limitation recites mental processes in addition to those identified in the rejection of the parent claim; thus, the claim recites a judicial exception.
The claim does not include any additional elements that amount to an integration of the judicial exception into a practical application, nor to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 12, the rejection of claim 1 is incorporated, and further, the claim recites: “wherein the inference service is operable to: receive an API request for use of the global model using feature data from a user received as part of the API request”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “send, to the user, an inference generated by the global model based on the feature data”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 13, the rejection of claim 1 is incorporated, and further, the claim recites: “wherein the inference service is operable to: receive a request from a user to use the global model”. Further, the claim recites: “send, to the user, an inference generated by the global model based on the feature data”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “provide access to the global model for downloading by the user”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II).
Regarding claim 14:
Step 1 Statutory Category: Claim 14 is directed to a method, which falls under one of the four statutory categories.
Step 2A Prong 1 Judicial exception: Claim 14 recites, in part, “identifying a group of clients to build a global model”. This limitation, under the broadest reasonable interpretation, covers the recitation of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion), in this case an observation. See MPEP § 2106.04(a)(2)(III). Further, the claim recites: “aggregating model parameters received from the group to build the global model, the model parameters being generated by the clients training a local ML model at their respective client sites using local data at their respective client sites”. This limitation, under the broadest reasonable interpretation, covers the recitation of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. A mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number”. See MPEP §2106.04(a)(2)(I)(C). Further, the claim recites: “calculating an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data”. This limitation, under the broadest reasonable interpretation, and in light of applicant’s specification paragraphs 0047-0048, covers the recitation of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. A mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number”. See MPEP §2106.04(a)(2)(I)(C).
Step 2A Prong 2 Integration into a practical application: This judicial exception is not integrated into a practical application. In particular the claim recites: “a global machine leaning (ML) model” and “by federated learning”. These limitations are additional elements that amount to generally linking the use of the judicial exception to a particular technological environment or field of use. See MPEP §2106.05(h). Further, the claim recites: “supplying a model definition for the global model to the group”. This limitation is an additional element that amounts adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g).
Step 2B Significantly more: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements: “a global machine leaning (ML) model” and “by federated learning” amount to generally linking the use of the judicial exception to a particular technological environment or field of use. Elements that merely amount to generally linking the use of the judicial exception to a particular technological environment or field of use cannot provide an inventive concept. Further, the additional element “supplying a model definition for the global model to the group” amounts adding insignificant extra-solution activity to the judicial exception, and further, is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 15, the rejection of claim 14 is incorporated, and further, the claim recites: “sending a request for training the global model to the clients of the group”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “receiving, from the clients and in response to the request for training, an indication that the clients want to join the group”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “storing registries that contain a model format and feature format for the global model”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to storing and retrieving information in memory which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “sending the model format and feature format to clients in the group as part of the model definition for use by each of the clients in training their local model”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 16, the rejection of claim 14 is incorporated, and further, claim 16 is substantially similar to claim 5 respectively, and is rejected in the same manner and reasoning applying.
Regarding claim 17, the rejection of claim 14 is incorporated, and further, claim 17 is substantially similar to claim 6 respectively, and is rejected in the same manner and reasoning applying.
Regarding claim 18:
Step 1 Statutory Category: Claim 18 is directed to an article of manufacture, which falls under one of the four statutory categories.
Step 2A Prong 1 Judicial exception: Claim 18 recites, in part, “identifying a group of clients to build a global model”. This limitation, under the broadest reasonable interpretation, covers the recitation of a mental process that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper (including an observation, evaluation, judgment, opinion), in this case an observation. See MPEP § 2106.04(a)(2)(III). Further, the claim recites: “aggregating model parameters received from the group to build the global model, the model parameters being generated by the clients training a local ML model at their respective client sites using local data at their respective client sites”. This limitation, under the broadest reasonable interpretation, covers the recitation of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. A mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number”. See MPEP §2106.04(a)(2)(I)(C). Further, the claim recites: “calculating an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data”. This limitation, under the broadest reasonable interpretation, and in light of applicant’s specification paragraphs 0047-0048, covers the recitation of a mathematical calculation, as directed to “a claim that recites a mathematical calculation, when the claim is given its broadest reasonable interpretation in light of the specification, will be considered as falling within the "mathematical concepts" grouping. A mathematical calculation is a mathematical operation (such as multiplication) or an act of calculating using mathematical methods to determine a variable or number”. See MPEP §2106.04(a)(2)(I)(C).
Step 2A Prong 2 Integration into a practical application: This judicial exception is not integrated into a practical application. In particular the claim recites: “A non-transitory computer readable storage media having instructions stored thereupon that, when executed by a processor of a computing system, the instructions cause the computing system to perform operations”. This limitation is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP §2106.05(f). Further, the claim recites: “a global machine leaning (ML) model” and “by federated learning”. These limitations are additional elements that amount to generally linking the use of the judicial exception to a particular technological environment or field of use. See MPEP §2106.05(h). Further, the claim recites: “supplying a model definition for the global model to the group”. This limitation is an additional element that amounts adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g).
Step 2B Significantly more: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional element: “A non-transitory computer readable storage media having instructions stored thereupon that, when executed by a processor of a computing system, the instructions cause the computing system to perform operations” amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. Elements that merely amount to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer in its ordinary capacity as a tool to perform an existing process cannot provide an inventive concept. Further, the additional elements: “a global machine leaning (ML) model” and “by federated learning” amount to generally linking the use of the judicial exception to a particular technological environment or field of use. Elements that merely amount to generally linking the use of the judicial exception to a particular technological environment or field of use cannot provide an inventive concept. Further, the additional element “supplying a model definition for the global model to the group” amounts adding insignificant extra-solution activity to the judicial exception, and further, is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 19, the rejection of claim 18 is incorporated, and further, the claim recites: “aggregating the model parameters from the clients based on at least one of client data set size of data used to train their respective local model and their local model’s accuracy”. This limitation recites mathematical concepts in addition to those identified in the rejection of the parent claim. Thus, the claim recites a judicial exception.
Further, the claim recites: “sending a request for training the global model to the clients of the group”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “receiving, from the clients and in response to the request for training, an indication that the clients want to join the group”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “storing registries that contain a model format and feature format for the global model”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to storing and retrieving information in memory which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). Further, the claim recites: “sending the model format and feature format to clients in the group as part of the model definition for use by each of the clients in training their local model”. This limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP §2106.05(g). Further, the limitation is directed to receiving or transmitting data over a network which courts have recognized as well-understood, routine, and conventional when they are claimed in a generic manner, see MPEP §2106.05(d)(II). The claim is not patent eligible.
Regarding claim 20, the rejection of claim 18 is incorporated, and further, claim 20 is substantially similar to claim 6 and claim 17 respectively, and is rejected in the same manner and reasoning applying.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 5-7, 10, 14, 16-18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Deng et al., AUCTION: Automated and Quality-Aware Client Selection Framework for Efficient Federated Learning, 12/24/2021, https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9647925, hereinafter referred to as “Deng”, in view of Nishio et al., Estimation of Individual Device Contributions for Incentivizing Federated Learning, 09/20/2020, https://arxiv.org/pdf/2009.09371 hereinafter referred to as “Nishio”.
Regarding claim 1, Deng teaches An architecture to build a global machine learning (ML) model (Deng, Page 1996, Abstract, Lines 4-6, “we propose AUCTION, an Automated and qUality aware Client selecTION framework for efficient FL, which can evaluate the learning quality of clients and select them automatically with quality-awareness for a given FL task within a limited budget”), the architecture comprising:
a platform to identify a group of clients to build a global model by federated learning (Deng, Page 1997, Col 1, Paragraph 3, Lines 1-8, “we consider an FL platform on the data market, where the machine learning model for different intelligent services is submitted to the platform with a budget … The functionality of the FL platform is to select a subset of clients to participate in distributed model training without exceeding the budget”), wherein the platform includes
a group manager to build the global model by supplying a model definition for the global model to the group (Deng, Page 1998, Section 2.1, Steps 1-3, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. 2) Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step. 3) Client selection. The FL platform conducts client selection to choose a subset of participants from the candidate clients, and then delivers the initial global model w0 to the selected participating clients”; Deng, Page 1998, Figure 1, the “FL task” can be seen being sent from the platform to the clients) and aggregating model parameters received from the group to build the global model (Deng, Page 1998, Section 2.1, Step 5, Global aggregation. The FL platform aggregates the received local model parameters from participating clients using the classical Federated Averaging algorithm”), the model parameters being generated by the clients training a local ML model at their respective client sites using local data at their respective client sites (Deng, Page 1998, Section 2.1, Step 4, “Local training. In each round r, based on the global model
w
r
, each participating client conducts model training individually by using the local data set
D
i
, the training results of which can be used to update the local model parameters
w
i
r
… After, local training, the updated local model parameters
w
i
r
of each participant are uploaded to the FL platform for the global model aggregation”).
Deng does not explicitly teach an incentive calculator communicably coupled to the group manager to calculate an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data.
Nishio teaches an incentive calculator communicably coupled to the group manager (Nishio, Pages 2-3, Section 3, Lines 10-13, “The evaluator estimates the individual contribution level of each client by comparing the ML models before and after being updated using the clients’ models”; see also Nishio, Page 3, Figure 1, the “Evaluator” can be seen connected to the “aggregator”) to calculate an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data (Nishio, Page 3, Section B, Step-wise contribution, Lines 1-5 and Equation 4, “We propose a light-weight but intuitive contribution estimation method based on step-wise contribution calculation. The metric used in the proposed method is defined as the sum of gains that include the client model in each round, calculated as:
G
i
S
W
C
=
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
∑
i
ϵ
C
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
”; Nishio, Page 3, Lines 3-4, “The evaluator sends back the rewards to each client based on its contribution”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the architecture of Deng to include an incentive calculator as taught by Nishio. The motivation to do so would have been that incentivizing clients motivates them to participate in federated learning and helps build a sustainable platform for federated learning (Nishio, Page 1, Abstract, Lines 4-6, “Appropriate incentive mechanisms that motivate the data and mobile-device owner to participate in FL is key to building a sustainable platform for FL”).
Regarding claim 5, the rejection of claim 1 is incorporated, and further, the proposed combination teaches wherein the group manager is operable to aggregate the model parameters from the clients based on client data set size of data used to train their respective local model (Deng, Page 1998, Section 2.1, Step 5, Lines 1-6, “Global aggregation. The FL platform aggregates the received local model parameters from participating clients using the classical Federated Averaging algorithm [7]
w
r
+
1
=
∑
i
=
1
N
d
i
w
i
r
∑
i
=
1
N
d
i
, where
d
i
=
|
D
i
|
is the number of data samples used by participating client
C
i
for local training”).
It is noted the claim recites alternative language and the reference teaches at least one of the alternatives.
Regarding claim 6, the rejection of claim 1 is incorporated, and further, the proposed combination teaches wherein the incentive calculator is operable to calculate the incentive based on an amount of each client’s contribution (Nishio, Page 3, Section B, Step-wise contribution, Lines 1-12 and Equation 4, “We propose a light-weight but intuitive contribution estimation method based on step-wise contribution calculation. The metric used in the proposed method is defined as the sum of gains that include the client model in each round, calculated as:
G
i
S
W
C
=
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
∑
i
ϵ
C
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
” where the denominator is used for normalization. The metric evaluates how much the client’s model improves the global model at each round and regards the sum of the step-wise contributions as the contribution of the FL client. This is based on the intuition that a client that improves model performance at each round will also contribute to the improvement of the final overall model performance; Nishio, Page 3, Lines 3-4, “The evaluator sends back the rewards to each client based on its contribution”).
Regarding claim 7, the rejection of claim 6 is incorporated, and further, the proposed combination teaches wherein each client’s contribution is based on their local model’s accuracy (Nishio, Page 3, Section B, Step-wise contribution, Lines 1-12 and Equation 4, “We propose a light-weight but intuitive contribution estimation method based on step-wise contribution calculation. The metric used in the proposed method is defined as the sum of gains that include the client model in each round, calculated as:
G
i
S
W
C
=
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
∑
i
ϵ
C
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
” where the denominator is used for normalization. The metric evaluates how much the client’s model improves the global model at each round and regards the sum of the step-wise contributions as the contribution of the FL client. This is based on the intuition that a client that improves model performance at each round will also contribute to the improvement of the final overall model performance; Nishio, Page 3, Lines 3-4, “The evaluator sends back the rewards to each client based on its contribution”; Nishio, Page 3, Section B, Paragraph 2, Lines 7-9, “P(M) is the performance score of model M, and the validation accuracy or loss can be used for the score”).
It is noted the claim recites alternative language, and the proposed combination teaches at least one of the alternatives.
Regarding claim 10, the rejection of claim 1 is incorporated, and further, the proposed combination teaches wherein the platform is operable to identify the group to build a global model in response to a request from one or more users of the global model (Deng, Page 1997, Col 1, Paragraph 3, Lines 1-8, “we consider an FL platform on the data market, where the machine learning model for different intelligent services is submitted to the platform with a budget … The functionality of the FL platform is to select a subset of clients to participate in distributed model training without exceeding the budget”; Deng, Page 1998, Section 2.1, Lines 6-8, “Users can submit FL tasks to the FL platform with a certain budget for recruiting clients to accomplish them”).
Regarding claim 14, Deng teaches A method for building a global machine learning (ML) model (Deng, Page 1996, Abstract, Lines 4-6, “we propose AUCTION, an Automated and qUality aware Client selecTION framework for efficient FL, which can evaluate the learning quality of clients and select them automatically with quality-awareness for a given FL task within a limited budget”), the method comprising:
identifying a group of clients to build a global model by federated learning (Deng, Page 1997, Col 1, Paragraph 3, Lines 1-8, “we consider an FL platform on the data market, where the machine learning model for different intelligent services is submitted to the platform with a budget … The functionality of the FL platform is to select a subset of clients to participate in distributed model training without exceeding the budget”);
supplying a model definition for the global model to the group (Deng, Page 1998, Section 2.1, Steps 1-3, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. 2) Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step. 3) Client selection. The FL platform conducts client selection to choose a subset of participants from the candidate clients, and then delivers the initial global model w0 to the selected participating clients”; Deng, Page 1998, Figure 1, the “FL task” can be seen being sent from the platform to the clients);
aggregating model parameters received from the group to build the global model (Deng, Page 1998, Section 2.1, Step 5, Global aggregation. The FL platform aggregates the received local model parameters from participating clients using the classical Federated Averaging algorithm”), the model parameters being generated by the clients training a local ML model at their respective client sites using local data at their respective client sites (Deng, Page 1998, Section 2.1, Step 4, “Local training. In each round r, based on the global model
w
r
, each participating client conducts model training individually by using the local data set
D
i
, the training results of which can be used to update the local model parameters
w
i
r
… After, local training, the updated local model parameters
w
i
r
of each participant are uploaded to the FL platform for the global model aggregation”).
Deng does not explicitly teach calculating an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data.
Nishio teaches calculating an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data (Nishio, Page 3, Section B, Step-wise contribution, Lines 1-5 and Equation 4, “We propose a light-weight but intuitive contribution estimation method based on step-wise contribution calculation. The metric used in the proposed method is defined as the sum of gains that include the client model in each round, calculated as:
G
i
S
W
C
=
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
∑
i
ϵ
C
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
”; Nishio, Page 3, Lines 3-4, “The evaluator sends back the rewards to each client based on its contribution”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the method of Deng to include calculating an incentive as taught by Nishio. The motivation to do so would have been that incentivizing clients motivates them to participate in federated learning and helps build a sustainable platform for federated learning (Nishio, Page 1, Abstract, Lines 4-6, “Appropriate incentive mechanisms that motivate the data and mobile-device owner to participate in FL is key to building a sustainable platform for FL”).
Regarding claim 16, the rejection of claim 14 is incorporated, and further, the claim recites: further comprising aggregating the model parameters from the clients based on client data set size of data used to train their respective local model (Deng, Page 1998, Section 2.1, Step 5, Lines 1-6, “Global aggregation. The FL platform aggregates the received local model parameters from participating clients using the classical Federated Averaging algorithm [7]
w
r
+
1
=
∑
i
=
1
N
d
i
w
i
r
∑
i
=
1
N
d
i
, where
d
i
=
|
D
i
|
is the number of data samples used by participating client
C
i
for local training”).
It is noted the claim recites alternative language and the reference teaches at least one of the alternatives.
Regarding claim 17, the rejection of claim 14 is incorporated, and further, the proposed combination teaches wherein calculating the incentive is based on an amount of each client’s contribution (Nishio, Page 3, Section B, Step-wise contribution, Lines 1-12 and Equation 4, “We propose a light-weight but intuitive contribution estimation method based on step-wise contribution calculation. The metric used in the proposed method is defined as the sum of gains that include the client model in each round, calculated as:
G
i
S
W
C
=
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
∑
i
ϵ
C
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
” where the denominator is used for normalization. The metric evaluates how much the client’s model improves the global model at each round and regards the sum of the step-wise contributions as the contribution of the FL client. This is based on the intuition that a client that improves model performance at each round will also contribute to the improvement of the final overall model performance; Nishio, Page 3, Lines 3-4, “The evaluator sends back the rewards to each client based on its contribution”).
Regarding claim 18, Deng teaches A non-transitory computer readable storage media having instructions stored thereupon that, when executed by a processor of a computing system, the instructions cause the computing system to perform operations (Deng, Page 2003, Section 6.1, Lines 1-9, “We build an experimental FL services market with real-world learning tasks to emulate mislabeled and non-IID scenarios and implement AUCTION as the client selection agent for the FL platform. We first emu late the entire FL process on the server under various scenarios where the emulated candidate clients have different training data size, mislabel rate, and data distribution, and train the client selection model of AUCTION offline through interacting with the experimental FL services market”; The “server” and experiments require the use of a generic computer which provides evidence for “a non-transitory computer readable storage media”, “instructions”, “a processor”, and “a computing system”) for building a global machine learning (ML) model (Deng, Page 1996, Abstract, Lines 4-6, “we propose AUCTION, an Automated and qUality aware Client selecTION framework for efficient FL, which can evaluate the learning quality of clients and select them automatically with quality-awareness for a given FL task within a limited budget”), the method comprising:
identifying a group of clients to build a global model by federated learning (Deng, Page 1997, Col 1, Paragraph 3, Lines 1-8, “we consider an FL platform on the data market, where the machine learning model for different intelligent services is submitted to the platform with a budget … The functionality of the FL platform is to select a subset of clients to participate in distributed model training without exceeding the budget”);
supplying a model definition for the global model to the group (Deng, Page 1998, Section 2.1, Steps 1-3, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. 2) Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step. 3) Client selection. The FL platform conducts client selection to choose a subset of participants from the candidate clients, and then delivers the initial global model w0 to the selected participating clients”; Deng, Page 1998, Figure 1, the “FL task” can be seen being sent from the platform to the clients);
aggregating model parameters received from the group to build the global model (Deng, Page 1998, Section 2.1, Step 5, Global aggregation. The FL platform aggregates the received local model parameters from participating clients using the classical Federated Averaging algorithm”), the model parameters being generated by the clients training a local ML model at their respective client sites using local data at their respective client sites (Deng, Page 1998, Section 2.1, Step 4, “Local training. In each round r, based on the global model
w
r
, each participating client conducts model training individually by using the local data set
D
i
, the training results of which can be used to update the local model parameters
w
i
r
… After, local training, the updated local model parameters
w
i
r
of each participant are uploaded to the FL platform for the global model aggregation”).
Deng does not explicitly teach calculating an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data.
Nishio teaches calculating an incentive to each client communicably coupled to the platform based on said each client’s contribution to train the global model, each client’s contribution including one or more model parameters generated as a result of training their local ML model with local data (Nishio, Page 3, Section B, Step-wise contribution, Lines 1-5 and Equation 4, “We propose a light-weight but intuitive contribution estimation method based on step-wise contribution calculation. The metric used in the proposed method is defined as the sum of gains that include the client model in each round, calculated as:
G
i
S
W
C
=
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
∑
i
ϵ
C
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
”; Nishio, Page 3, Lines 3-4, “The evaluator sends back the rewards to each client based on its contribution”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the method of Deng to include calculating an incentive as taught by Nishio. The motivation to do so would have been that incentivizing clients motivates them to participate in federated learning and helps build a sustainable platform for federated learning (Nishio, Page 1, Abstract, Lines 4-6, “Appropriate incentive mechanisms that motivate the data and mobile-device owner to participate in FL is key to building a sustainable platform for FL”).
Regarding claim 20, the rejection of claim 18 is incorporated, and further, the proposed combination teaches wherein calculating the incentive is based on an amount of each client’s contribution (Nishio, Page 3, Section B, Step-wise contribution, Lines 1-12 and Equation 4, “We propose a light-weight but intuitive contribution estimation method based on step-wise contribution calculation. The metric used in the proposed method is defined as the sum of gains that include the client model in each round, calculated as:
G
i
S
W
C
=
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
∑
i
ϵ
C
∑
r
=
1
r
e
n
d
P
M
C
r
-
P
M
C
{
i
}
r
” where the denominator is used for normalization. The metric evaluates how much the client’s model improves the global model at each round and regards the sum of the step-wise contributions as the contribution of the FL client. This is based on the intuition that a client that improves model performance at each round will also contribute to the improvement of the final overall model performance; Nishio, Page 3, Lines 3-4, “The evaluator sends back the rewards to each client based on its contribution”).
Claims 2-4, 11-13, 15, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Deng in view of Nishio in further view of Lamoureux et al., U.S. Patent Application Publication No. 20250299106, hereinafter referred to as “Lamoureux”.
Regarding claim 2, the rejection of claim 1 is incorporated, and further, the proposed combination teaches a server to provide … the model definition (Deng, Page 2003, Section 6.1, Lines 4-9, “We first emu late the entire FL process on the server under various scenarios where the emulated candidate clients have different training data size, mislabel rate, and data distribution, and train the client selection model of AUCTION offline through interacting with the experimental FL services market”; Deng, Page 1998, Section 2.1, Step 1, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results; see also Deng, Page 1998, Figure 1, the “FL platform” is shown sending the “FL task” to the clients).
The proposed combination does not explicitly teach registries for storing a model format and feature format for the global model nor the model format and feature format to the group as part of the model definition.
Lamoureux teaches registries for storing a model format and feature format for the global model and the model format and feature format to the group as part of the model definition (Lamoureux, Paragraph 0108, Lines 5-10, “the machine learning system may retrieve, from the model registry, the model definition (e.g., architecture and hyperparameters, input features, and the like) and configuration information (e.g., preprocessing operations, data storage locations, deployment type, and the like) for the model indicated in the request”; The “architecture and hyperparameters” are considered to be the “model format” and the “input features” are considered to be the “feature format”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the architecture as taught by the proposed combination to include registries for storing a model format and feature format which are part of the model definition as taught by Lamoureux. The motivation to do so would have been that storing the model and feature formats in registries allows the models to be stored with indicators to show whether the model is ready for training or deployment (Lamoureux, Paragraphs 0054-0055).
Regarding claim 3, the rejection of claim 2 is incorporated, and further, the proposed combination teaches wherein the server provides the model and feature format to the group as part of the model definition (Lamoureux, Paragraph 0108, Lines 5-10, “the machine learning system may retrieve, from the model registry, the model definition (e.g., architecture and hyperparameters, input features, and the like) and configuration information (e.g., preprocessing operations, data storage locations, deployment type, and the like) for the model indicated in the request”; The “architecture and hyperparameters” are considered to be the “model format” and the “input features” are considered to be the “feature format”) after the group manager sends a request for training to the clients and the clients join the group in response to the request for training (Deng, Page 1998, Section 2.1, Steps 1-3, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. 2) Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step. 3) Client selection. The FL platform conducts client selection to choose a subset of participants from the candidate clients, and then delivers the initial global model w0 to the selected participating clients”; Deng, Page 1998, Figure 1).
Regarding claim 4, the rejection of claim 2 is incorporated, and further, the proposed combination teaches wherein the server provides the model format and feature format (Lamoureux, Paragraph 0108, Lines 5-10, “the machine learning system may retrieve, from the model registry, the model definition (e.g., architecture and hyperparameters, input features, and the like) and configuration information (e.g., preprocessing operations, data storage locations, deployment type, and the like) for the model indicated in the request”; The “architecture and hyperparameters” are considered to be the “model format” and the “input features” are considered to be the “feature format”) to said each client for use when training their local model (Deng, Page 1998, Section 2.1, Steps 1-3, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. 2) Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step. 3) Client selection. The FL platform conducts client selection to choose a subset of participants from the candidate clients, and then delivers the initial global model w0 to the selected participating clients”; Deng, Page 1998, Figure 1).
Regarding claim 11, the rejection of claim 1 is incorporated.
The proposed combination thus far does not explicitly teach an inference service responsive to request for use of the global model by one or more users.
Lamoureux teaches an inference service responsive to request for use of the global model by one or more users (Lamoureux, Paragraph 0057, Lines 2-5, “The serving component 230 can generally access the definitions and configurations in the model registry 220 to instantiate pipelines 235, 240, and/or 245”; Lamoureux, Paragraph 0059, Lines 1-4, “the real-time inference pipeline 235 includes a copy or instance of the model 250A, as well as an API 255 that can be used to enable or provide access to the model 250A (e.g., to application(s) 270A)”; Lamoureux, Page 0114, Lines 3-5, “using an API (e.g., API 255 of FIG. 2), the requesting entity (which may be an automated application, a user-controlled application, and the like)”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the architecture of the proposed combination to include an inferencing service as taught by Lamoureux. The motivation to do so would have been that users can use the inferencing service to produce output for their input samples, and optionally create new training samples (Lamoureux, Paragraph 0101).
Regarding claim 12, the rejection of claim 1 is incorporated, and further, the proposed combination thus far does not explicitly teach wherein the inference service is operable to: receive an API request for use of the global model using feature data from a user received as part of the API request; and send, to the user, an inference generated by the global model based on the feature data.
Lamoureux teaches wherein the inference service is operable to: receive an API request for use of the global model using feature data from a user received as part of the API request (Lamoureux, Paragraph 0059, Lines 1-6, “the real-time inference pipeline 235 includes a copy or instance of the model 250A, as well as an API 255 that can be used to enable or provide access to the model 250A (e.g., to application(s) 270A). For example, the application 270A may use the API 255 to provide input data to the real-time inference pipeline 235”; Lamoureux, Page 0114, Lines 3-5, “using an API (e.g., API 255 of FIG. 2), the requesting entity (which may be an automated application, a user-controlled application, and the like)”); and
send, to the user, an inference generated by the global model based on the feature data (Lamoureux, Paragraph 0059, Lines 5-9, “the application 270A may use the API 255 to provide input data to the real-time inference pipeline 235, which then processes it with the model 250A to generate an output inference. This output can then be returned, via the API 255, back to the application 270”; Lamoureux, Page 0114, Lines 3-5, “using an API (e.g., API 255 of FIG. 2), the requesting entity (which may be an automated application, a user-controlled application, and the like)”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the architecture of the proposed combination to include an inferencing service as taught by Lamoureux. The motivation to do so would have been that users can use the inferencing service to produce output for their input samples, and optionally create new training samples (Lamoureux, Paragraph 0101).
Regarding claim 13, the rejection of claim 1 is incorporated, and further, the proposed combination thus far does not explicitly teach wherein the inference service is operable to: receive a request from a user to use the global model; and provide access to the global model for downloading by the user.
Lamoureux teaches wherein the inference service is operable to: receive a request from a user to use the global model; and provide access to the global model for downloading by the user (Lamoureux, Paragraph 0059, Lines 1-4, “the real-time inference pipeline 235 includes a copy or instance of the model 250A, as well as an API 255 that can be used to enable or provide access to the model 250A (e.g., to application(s) 270A)”; Lamoureux, Page 0114, Lines 3-5, “using an API (e.g., API 255 of FIG. 2), the requesting entity (which may be an automated application, a user-controlled application, and the like)”).
Regarding claim 15, the rejection of claim 14 is incorporated, and further, the proposed combination teaches
sending a request for training the global model to the clients of the group; receiving, from the clients and in response to the request for training, an indication that the clients want to join the group (Deng, Page 1998, Section 2.1, Steps 1-2, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. 2) Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step; see also Deng, Page 1998, Figure 1, the “FL task can be seen being sent from the platform to the clients)
…
sending … the model definition for use by each of the clients in training their local model (Deng, Page 2003, Section 6.1, Lines 4-9, “We first emu late the entire FL process on the server under various scenarios where the emulated candidate clients have different training data size, mislabel rate, and data distribution, and train the client selection model of AUCTION offline through interacting with the experimental FL services market”; Deng, Page 1998, Section 2.1, Step 1, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results; see also Deng, Page 1998, Figure 1, the “FL platform” is shown sending the “FL task” to the clients).
The proposed combination thus far does not explicitly teach storing registries that contain a model format and feature format for the global model nor the model format and feature format to clients in the group as part of the model definition.
Lamoureux teaches storing registries that contain a model format and feature format for the global model and the model format and feature format to clients in the group as part of the model definition (Lamoureux, Paragraph 0108, Lines 5-10, “the machine learning system may retrieve, from the model registry, the model definition (e.g., architecture and hyperparameters, input features, and the like) and configuration information (e.g., preprocessing operations, data storage locations, deployment type, and the like) for the model indicated in the request”; The “architecture and hyperparameters” are considered to be the “model format” and the “input features” are considered to be the “feature format”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the architecture as taught by the proposed combination to include registries for storing a model format and feature format which are part of the model definition as taught by Lamoureux. The motivation to do so would have been that storing the model and feature formats in registries allows the models to be stored with indicators to show whether the model is ready for training or deployment (Lamoureux, Paragraphs 0054-0055).
Regarding claim 19, the rejection of claim 18 is incorporated, and further, the proposed combination teaches sending a request for training the global model to the clients of the group; receiving, from the clients and in response to the request for training, an indication that the clients want to join the group (Deng, Page 1998, Section 2.1, Steps 1-2, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. 2) Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step; see also Deng, Page 1998, Figure 1, the “FL task can be seen being sent from the platform to the clients)
…
sending … the model definition for use by each of the clients in training their local model (Deng, Page 2003, Section 6.1, Lines 4-9, “We first emu late the entire FL process on the server under various scenarios where the emulated candidate clients have different training data size, mislabel rate, and data distribution, and train the client selection model of AUCTION offline through interacting with the experimental FL services market”; Deng, Page 1998, Section 2.1, Step 1, “1) Task initialization. A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results; see also Deng, Page 1998, Figure 1, the “FL platform” is shown sending the “FL task” to the clients); and
aggregating the model parameters from the clients based on at least one of client data set size of data used to train their respective local model and their local model’s accuracy (Deng, Page 1998, Section 2.1, Step 5, Lines 1-6, “Global aggregation. The FL platform aggregates the received local model parameters from participating clients using the classical Federated Averaging algorithm [7]
w
r
+
1
=
∑
i
=
1
N
d
i
w
i
r
∑
i
=
1
N
d
i
, where
d
i
=
|
D
i
|
is the number of data samples used by participating client
C
i
for local training”).
The proposed combination thus far does not explicitly teach storing registries that contain a model format and feature format for the global model nor the model format and feature format to clients in the group as part of the model definition.
Lamoureux teaches storing registries that contain a model format and feature format for the global model and the model format and feature format to clients in the group as part of the model definition (Lamoureux, Paragraph 0108, Lines 5-10, “the machine learning system may retrieve, from the model registry, the model definition (e.g., architecture and hyperparameters, input features, and the like) and configuration information (e.g., preprocessing operations, data storage locations, deployment type, and the like) for the model indicated in the request”; The “architecture and hyperparameters” are considered to be the “model format” and the “input features” are considered to be the “feature format”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to have modified the architecture as taught by the proposed combination to include registries for storing a model format and feature format which are part of the model definition as taught by Lamoureux. The motivation to do so would have been that storing the model and feature formats in registries allows the models to be stored with indicators to show whether the model is ready for training or deployment (Lamoureux, Paragraphs 0054-0055).
Claims 8-9 are rejected under 35 U.S.C. 103 as being unpatentable over Deng in view of Nishio in further view of Zhang et al., Auction-Based Ex-Post-Payment Incentive Mechanism Design for Horizontal Federated Learning with Reputation and Contribution Measurement, 03/15/2022, https://arxiv.org/pdf/2201.02410, hereinafter referred to as “Zhang”.
Regarding claim 8, the rejection of claim 6 is incorporated, and further, the proposed combination thus far does not explicitly teach wherein the incentive calculator calculates the incentive based on stored training and trading histories.
Zhang teaches wherein the incentive calculator calculates the incentive based on stored training and trading histories (Zhang, Page 3, Section 4.2, Lines 1-8, “Reputation is a rating of the quality and reliability of workers, allowing publishers to select high-quality workers. First, we model the reputation of a worker in a certain task as internal reputation and then integrate the reputation in all historical tasks as the accumulated reputation. The internal reputation represents the performance of the worker in the current task, which is related to his contribution and the quality detection of his local model”).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention to have modified the architecture of the proposed combination to include calculating the incentive based on stored training and trading histories as taught by Zhang. The motivation to do so would have been the ability to select high-quality workers with good quality and reliability (Zhang, Page 3, Section 4.2, Lines 1-2, “Reputation is a rating of the quality and reliability of workers, allowing publishers to select high-quality workers”).
Regarding claim 9, the rejection of claim 1 is incorporated, and further, the proposed combination thus far does not explicitly teach wherein the local model trained by the clients has identical feature sets.
Zhang teaches wherein the local model trained by the clients has identical feature sets (Zhang, Page 1, Abstract, Lines 10-13, “Therefore, we design an auction-based incentive mechanism for horizontal federated learning with reputation and contribution measurement”; A person of ordinary skill in the art would recognize that “horizontal federated learning” requires clients with identical feature sets).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the invention, to include modifying the architecture of the proposed combination to include the local models to have identical feature sets as taught by Zhang. The motivation to do so would have been that horizontal federated learning makes full use of everyone’s storage and computing capabilities, satisfies privacy protection, and can be used in banking, healthcare, transportation, etc., (Zhang, Page 1, Section 1).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Kong et al., INCENTIVIZING FEDERATED LEARNING, 05/22/2022, https://arxiv.org/pdf/2205.10951 teaches a non-monetary incentive mechanism that uses model performance as a reward, the server evaluates the performance of every client’s uploaded model in each round and distributes different models to clients based on the evaluation result.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MOLLY CLARKE SIPPEL whose telephone number is (571)272-3270. The examiner can normally be reached Monday - Friday, 7:30 a.m. - 4:30 p.m. ET..
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, Kakali Chaki can be reached at (571)272-3719. 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.
/M.C.S./ Examiner, Art Unit 2122
/MICHAEL H HOANG/ PRIMARY EXAMINER, Art Unit 2122
Prosecution Insights