DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This action is responsive to the original application filed on 4/11/2023. Acknowledgment is made with respect to a claim of priority to Chinese Application CN202210399999.6 filed on 4/15/2022.
Claim Objections
Applicant is advised that should claim 4 be found allowable, claim 6 will be objected to under 37 CFR 1.75 as being a substantial duplicate thereof. When two claims in an application are duplicates or else are so close in content that they both cover the same thing, despite a slight difference in wording, it is proper after allowing one claim to object to the other as being a substantial duplicate of the allowed claim. See MPEP § 608.01(m).
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. The analysis of the claims will follow the 2019 Revised Patent Subject Matter Eligibility Guidance, 84 Fed. Reg. 50 (“2019 PEG”).
When considering subject matter eligibility under 35 U.S.C. 101, it must be determined whether the claim is directed to one of the four statutory categories of invention, i.e., process, machine, manufacture, or composition of matter (Step 1). If the claim does fall within one of the statutory categories, the second step in the analysis is to determine whether the claim is directed to a judicial exception (Step 2A). The Step 2A analysis is broken into two prongs. In the first prong (Step 2A, Prong 1), it is determined whether or not the claims recite a judicial exception (e.g., mathematical concepts, mental processes, certain methods of organizing human activity). If it is determined in Step 2A, Prong 1 that the claims recite a judicial exception, the analysis proceeds to the second prong (Step 2A, Prong 2), where it is determined whether or not the claims integrate the judicial exception into a practical application. If it is determined at step 2A, Prong 2 that the claims do not integrate the judicial exception into a practical application, the analysis proceeds to determining whether the claim is a patent-eligible application of the exception (Step 2B). If an abstract idea is present in the claim, any element or combination of elements in the claim must be sufficient to ensure that the claim integrates the judicial exception into a practical application, or else amounts to significantly more than the abstract idea itself.
Claim 1
Step 1: The claim recites a method; therefore, it is directed to the statutory category of a process.
Step 2A Prong 1: The claim recites, inter alia:
determining, based on the data volume information and the node local gradient, a global gradient of a federated learning model that the participating node participates in: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a global gradient of a model based on data volume information and a local gradient, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. For example, one can practically and mentally determine a gradient based on an amount of data or data volume and another gradient.
determining, based on the node local gradient of the participating node and the global gradient, a degree of participation of the participating node, wherein the degree of participation indicates a degree of participation of the participating node in federated learning model training: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mathematical concept of determining a degree of participation for a participating node, which is performed through mathematical computation as evidenced by paragraph [0056] of the originally filed specification.
determining, based on the degree of participation, an actual model gradient of the participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a gradient based on a degree of participation for a node, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. For example, one can practically and mentally determine a gradient or change in information based on an amount of participation or activity for a node.
Step 2A Prong 2: The claim does not recite any additional limitations which integrate the abstract idea into a practical application. Specifically, the additional elements consist of “obtaining data volume information of a participating node, wherein the data volume information indicates an amount of data used by the participating node to train, based on local data, a basic training model, and wherein the local data comprises user data of a target organization corresponding to the participating node” and “obtaining, based on the local data and by the participating node, a node local gradient by training the basic training model”.
The additional element of “obtaining, based on the local data and by the participating node, a node local gradient by training the basic training model” amount to reciting only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished because it is not clear how the node local gradient is obtained through a broadly claimed training procedure of a basic training model. Thus, the additional elements amount to no more than a recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer (see MPEP § 2106.05(f)).
The additional element “obtaining data volume information of a participating node, wherein the data volume information indicates an amount of data used by the participating node to train, based on local data, a basic training model, and wherein the local data comprises user data of a target organization corresponding to the participating node” is insignificant extra-solution activity required for any uses of the abstract ideas (see MPEP § 2106.05(g)).
Thus, even when viewed individually and as an ordered combination, these additional elements do not integrate the abstract idea into a practical application and the claim is thus directed to the abstract idea.
Step 2B: Finally, the claim taken as a whole does not contain an inventive concept which provides significantly more than the abstract idea.
The additional element of “obtaining, based on the local data and by the participating node, a node local gradient by training the basic training model” amount to reciting only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished because it is not clear how the node local gradient is obtained through a broadly claimed training procedure of a basic training model. Thus, the additional elements amount to no more than a recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer (see MPEP § 2106.05(f)).
The additional element “obtaining data volume information of a participating node, wherein the data volume information indicates an amount of data used by the participating node to train, based on local data, a basic training model, and wherein the local data comprises user data of a target organization corresponding to the participating node” is insignificant extra-solution activity required for any uses of the abstract ideas (see MPEP § 2106.05(g)), and is a well-understood, routine, conventional activity (see MPEP § 2106.05(d)(II)(i); “Receiving or transmitting data over a network”).
Taken alone or in combination, the additional elements of the claim do not provide an inventive concept and thus the claim is subject-matter ineligible.
Claim 2
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
obtaining a marginal loss of the participating node, wherein the marginal loss represents a degree of influence of the node local gradient of the participating node on performance of the federated learning model: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mathematical concept of obtaining a marginal loss of a node, which is performed through mathematical computation as evidenced by paragraph [0043] of the originally filed specification.
determining, based on the marginal loss, node mass of the participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mathematical concept of determining a node mass for a node, which is performed through mathematical computation as evidenced by paragraph [0047] of the originally filed specification.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 3
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining, based on the data volume information, the node local gradient, and the node mass, the global gradient of the federated learning model that the participating node participates in: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mathematical concept of determining a global gradient, which is performed through mathematical computation as evidenced by paragraph [0057] of the originally filed specification.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 4
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of the preceding claims from which it depends.
Step 2A Prong 2, Step 2B: The additional element of “wherein the participating node participating in the federated learning model comprises a plurality of participating nodes” amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). Taken alone or in combination, the additional elements of the claim do not provide an inventive concept, integrate the abstract ideas into a practical application, or provide significantly more than the abstract ideas of the claim and thus the claim is subject-matter ineligible.
Claim 5
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining, based on a node local gradient of each participating node in the plurality of participating nodes, a first reference global model: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a reference global model, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. For example, one can practically and mentally determine a model based on gradients.
determining, based on a node local gradient of each participating node other than the participating node in the plurality of participating nodes, a second reference global model: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a reference global model, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. For example, one can practically and mentally determine a model based on gradients.
determining, based on a predetermined verification set, a first model loss of the first reference global model: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a model loss, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining, based on the predetermined verification set, a second model loss of the second reference global model: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a model loss, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining, based on the first model loss and the second model loss, the marginal loss of the participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a marginal loss based on model losses, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 6
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of the preceding claims from which it depends.
Step 2A Prong 2, Step 2B: The additional element of “wherein the participating node participating in the federated learning model comprises a plurality of participating nodes” amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). Taken alone or in combination, the additional elements of the claim do not provide an inventive concept, integrate the abstract ideas into a practical application, or provide significantly more than the abstract ideas of the claim and thus the claim is subject-matter ineligible.
Claim 7
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining, based on a marginal loss of each participating node in the plurality of participating nodes and a normalization algorithm, the node mass of the participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mathematical concept of determining a node mass for a node, which is performed through mathematical computation as evidenced by paragraph [0047] of the originally filed specification.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 8
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining a participating node with a marginal loss greater than or equal to a predetermined loss threshold in the plurality of participating nodes as an effective participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a an effective node based on a marginal loss, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The additional element of “wherein the participating node participating in the federated learning model comprises a plurality of participating nodes” amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). Taken alone or in combination, the additional elements of the claim do not provide an inventive concept, integrate the abstract ideas into a practical application, or provide significantly more than the abstract ideas of the claim and thus the claim is subject-matter ineligible.
Claim 9
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
performing, based on data volume information of the participating node and node mass of the participating node and to obtain the global gradient, an aggregation operation on a node local gradient of each of the effective participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of performing an aggregation operation on node gradients, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 10
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of the preceding claims from which it depends.
Step 2A Prong 2, Step 2B: The additional element of “wherein the participating node participating in the federated learning model comprises a plurality of participating nodes” amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). Taken alone or in combination, the additional elements of the claim do not provide an inventive concept, integrate the abstract ideas into a practical application, or provide significantly more than the abstract ideas of the claim and thus the claim is subject-matter ineligible.
Claim 11
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining a node contribution degree of each of the plurality of participating nodes based on the node local gradient of the participating node and the global gradient: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining contribution degrees of nodes based on gradients, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining a relative contribution degree of the participating node based on a node contribution degree of the participating node and the node contribution degree of each participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining contribution degrees of nodes based on other contribution degrees, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 12
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
obtaining a trustworthiness parameter of the participating node, wherein the trustworthiness parameter represents a comprehensive degree of reliability of the participating node in the target quantity of rounds of iterative calculations of the global gradient: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of obtaining trustworthiness parameters, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining a reputation degree of the participating node based on the relative contribution degree and the trustworthiness parameter: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining reputation degrees based on another degree and a trustworthiness parameter, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The additional element of “wherein the global gradient comprises a gradient obtained through a target quantity of rounds of iterative calculations” amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). Taken alone or in combination, the additional elements of the claim do not provide an inventive concept, integrate the abstract ideas into a practical application, or provide significantly more than the abstract ideas of the claim and thus the claim is subject-matter ineligible.
Claim 13
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining a first quantity of times that the participating node is determined as an effective participating node in the target quantity of rounds of iterative calculations, wherein the effective participating node represents a participating node with a marginal loss greater than or equal to a predetermined loss threshold: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a number of times that the node is effective, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining a second quantity of times that the participating node is determined as an ineffective participating node in the target quantity of rounds of iterative calculations, wherein the ineffective participating node represents a participating node with a marginal loss less than the predetermined loss threshold: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a number of times that the node is ineffective, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining the trustworthiness parameter of the participating node based on the first quantity of times and the second quantity of times: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining trustworthiness based determining a number of times that the node is ineffective or effective, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 14
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the abstract ideas of the preceding claims from which it depends.
Step 2A Prong 2, Step 2B: The additional element of “wherein the global gradient comprises a plurality of global gradient factors, wherein the participating node participating in the federated learning model comprises a plurality of participating nodes, and wherein the global gradient comprises the gradient obtained through the target quantity of rounds of iterative calculations” amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). Taken alone or in combination, the additional elements of the claim do not provide an inventive concept, integrate the abstract ideas into a practical application, or provide significantly more than the abstract ideas of the claim and thus the claim is subject-matter ineligible.
Claim 15
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining a quantity of matching gradients corresponding to the participating node based on a ratio of the reputation degree of the participating node to a greatest reputation degree, wherein the greatest reputation degree represents the greatest reputation degree in reputation degrees of the plurality of participating nodes;: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a number of matching gradients, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
selecting global gradient factors of the quantity of matching gradients from the global gradient to obtain the actual model gradient: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of selecting gradient factors, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 16
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
obtaining a node influence degree of each global gradient factor relative to the participating node, wherein the node influence degree of each global gradient factor indicates a degree that each global gradient factor is influenced by the participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of obtaining a node influence degree, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
sorting, based on the node influence degree of each global gradient factor and to obtain sorted global gradient factors, the global gradient factors in the global gradient: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of sorting gradient factors, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The additional element of “wherein the global gradient comprises a plurality of global gradient factors, wherein the node local gradient of the participating node comprises a plurality of local gradient factors” amounts to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). Taken alone or in combination, the additional elements of the claim do not provide an inventive concept, integrate the abstract ideas into a practical application, or provide significantly more than the abstract ideas of the claim and thus the claim is subject-matter ineligible.
Claim 17
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
selecting, based on a predetermined order and to obtain the actual model gradient, global gradient factors of the quantity of matching gradients from the sorted global gradient factors: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of selecting gradient factors, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 18
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
determining a first distribution parameter corresponding to each local gradient factor in the node local gradient of the participating node, wherein the first distribution parameter indicates a proportion of each local gradient factor in the node local gradient of the participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a distribution parameter, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining a second distribution parameter corresponding to each global gradient factor in the global gradient, wherein the second distribution parameter indicates a proportion of each global gradient factor in the global gradient: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a distribution parameter, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
determining, based on the first distribution parameter and the second distribution parameter, the node influence degree of each global gradient factor relative to the participating node: Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining a node influence degree, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper.
Step 2A Prong 2, Step 2B: The claim does not recite any additional elements that are sufficient to integrate the judicial exceptions into a practical application or amount to significantly more than the judicial exception. As such, the claim is ineligible.
Claim 19
Claim 19 recites a non-transitory computer readable medium (step 1: a manufacture) using a computer system to perform the steps of claim 1, which by MPEP 2106.05(f) (“apply it”) cannot integrate an abstract idea into a practical application or provide significantly more than the abstract idea by itself, and is thus rejected for the same reasons set forth in the rejection of claim 1.
Claim 20
Claim 20 recites a system (step 1: a machine) using a computer and computer memory to perform the steps of claim 1, which by MPEP 2106.05(f) (“apply it”) cannot integrate an abstract idea into a practical application or provide significantly more than the abstract idea by itself, and is thus rejected for the same reasons set forth in the rejection of claim 1.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-20 are rejected under 35 USC § 102(a)(1) as being anticipated by Gao et al. (Gao et al., “FIFL: A Fair Incentive Mechanism for Federated Learning”, Oct. 5, 2021, ICPP '21: Proceedings of the 50th International Conference on Parallel Processing, Article No.: 82, pp. 1-10, hereinafter “Gao”).
Regarding claim 1, Gao discloses [a] computer-implemented method for model gradient determination based on federated learning, comprising: (Abstract; “we propose FIFL, a fair incentive mechanism for federated learning. FIFL rewards workers fairly to attract reliable and efficient ones while punishing and eliminating the malicious ones based on a dynamic real-time worker assessment mechanism”; and §3.1)
obtaining data volume information of a participating node, wherein the data volume information indicates an amount of data used by the participating node to train, based on local data, a basic training model, and wherein the local data comprises user data of a target organization corresponding to the participating node; (§3.1; “there are N workers, the local training data on worker i is Di, which contains ni training samples”, wherein the section discloses obtaining data volume information “ni” of a participating node or worker, and this data indicates an amount of data used by the node or worker to train a basic learning model and wherein the local or worker data comprises data of a target organization corresponding to the node or worker; and §5.1; “We implement FIFL based on PyTorch 1 and build a simulator consisting of different agents with different data qualities in order to emulate the corresponding participating workers”, which discloses local worker data of an organization or entity corresponding to the node or worker; and §5.2; “The amount of local training samples for each worker is randomly generated, which is uniformly distributed in [1,10000]. We divide workers into ten groups according to their qualities, where the i-th group’s worker owns samples in the range of [1000∗(i−1),1000∗(i)].”; and §5.3; “We build a federated learning system consisting of 10 workers. The training data are uniformly distributed to each worker. We test the performance of the global model under various scenarios. Specifically, for MNIST data set, each of the 10 workers is given 6000 samples randomly sampled from the training set; for the CIFAR10 data set, each worker has 5000 samples. We use LeNet for MNIST and ResNet for CIFAR10”)
obtaining, based on the local data and by the participating node, a node local gradient by training the basic training model; (§3.1; “In the training process of FL, at the beginning of local training iteration t each worker sets its local parameter as the global parameter θi,0 = θt, and then computes its local gradients as: Gi = K k=1 ∂Li(θi,k,Di) ∂θi,k , where K is local training iterations. After that, workers upload their local gradients to the server. The server aggregates local gradients to get the global gradient as follows: N ˜ G = where ni N i=1 ( ni N j=1 nj Gi) ”, which discloses, under a broadest reasonable interpretation of the claim language, obtaining or determining a local gradient Gi by training the basic learning model, and this is based on local data)
determining, based on the data volume information and the node local gradient, a global gradient of a federated learning model that the participating node participates in; (§3.1; “Assuming there are N workers, the local training data on worker i is Di, which contains ni training samples … In the training process of FL, at the beginning of local training iteration t each worker sets its local parameter as the global parameter θi,0 = θt, and then computes its local gradients as: Gi = K k=1 ∂Li(θi,k,Di) ∂θi,k , where K is local training iterations. After that, workers upload their local gradients to the server. The server aggregates local gradients to get the global gradient as follows: N ˜ G = where ni N i=1 ( ni N j=1 nj Gi)”, which discloses determining the global gradient of the federated ML model)
determining, based on the node local gradient of the participating node and the global gradient, a degree of participation of the participating node, wherein the degree of participation indicates a degree of participation of the participating node in federated learning model training; and (§1; “We define workers’ contributions based on the deviation distance of workers’ local gradients to the global gradient to distinguish their utilities in real time”, wherein the degree of participation is interpreted as the deviation of distance of worker’s local gradients to the global gradient; and §4.3; “The contribution module aims to measure the utilities of workers. We first introduce the theoretical basis for using gradient similarity as a contribution measure”, wherein the contribution measure is interpreted as the degree of participation)
determining, based on the degree of participation, an actual model gradient of the participating node (§4.3; “If we take θ +Gi as y, take the fixed θ + ˜G as x, and take the function L(F()) as L, then both the upper bound and the lower bound of L(F(θ + Gi)) are positive correlation terms of ||Gi − ˜G||2. That indicates that high quality workers’ local gradients are close to ˜ G, while the low-quality workers’ gradients have more significant deviations from the ˜G”, wherein the actual model gradient is interpreted as “˜G” and it is based on a degree of participation or contribution measure; and §4.1; “Based on Equation 7, servers identify and reject attackers’ gradients, the aggregated global gradient becomes ˜G = N i=1( niri N j=0 rinj Gi). The hyperparameter Sy is the boundary to distinguish attackers from workers. Sy controls the trade-off between the detection accuracy and the false alarm rate in detecting attackers”).
Regarding claim 19, it is a non-transitory computer-readable medium claim corresponding to the steps of claim 1, and is rejected for the same reasons as claim 1.
Regarding claim 20, it is a system claim corresponding to the steps of claim 1, and is rejected for the same reasons as claim 1.
Regarding claim 2, the rejection of claim 1 is incorporated and Gao further discloses obtaining a marginal loss of the participating node, wherein the marginal loss represents a degree of influence of the node local gradient of the participating node on performance of the federated learning model; and determining, based on the marginal loss, node mass of the participating node (§4.2; “we calculate workers’ contribution based on the distance between the local gradient and global gradient. Specifically, we calculate the gradient distance for worker i as follows: M bi = Dis( ˜G,Gi) = j=1 Dis( ˜дj ,дj i ), (13) where Dis() is square of the Euclidean norm || ˜G,Gi||2”, wherein the Euclidean norm is a loss and bi is a marginal loss).
Regarding claim 3, the rejection of claims 1 and 2 are incorporated and Gao further discloses determining, based on the data volume information, the node local gradient, and the node mass, the global gradient of the federated learning model that the participating node participates in (§3.1; “In the training process of FL, at the beginning of local training iteration t each worker sets its local parameter as the global parameter θi,0 = θt, and then computes its local gradients as: Gi = K k=1 ∂Li(θi,k,Di) ∂θi,k , where K is local training iterations”; and §4.4; “Figure 3 abstractly expresses the role of contribution, reputation and reward in FIFL. Contribution is the indicator to measure workers’ utilities in each iteration. Reputation is the indicator that reflects the workers’ trustworthiness of producing useful gradients. The task publisher determines the reward share as the product of contribution and reputation”).
Regarding claim 4, the rejection of claims 1, 2, and 3 are incorporated and Gao further discloses wherein the participating node participating in the federated learning model comprises a plurality of participating nodes (Figures 1 and 3; the figures disclose a plurality of participating nodes or workers).
Regarding claim 5, the rejection of claims 1-4 are incorporated and Gao further discloses determining, based on a node local gradient of each participating node in the plurality of participating nodes, a first reference global model; determining, based on a node local gradient of each participating node other than the participating node in the plurality of participating nodes, a second reference global model; determining, based on a predetermined verification set, a first model loss of the first reference global model; determining, based on the predetermined verification set, a second model loss of the second reference global model; and determining, based on the first model loss and the second model loss, the marginal loss of the participating node (§4.2; “we calculate workers’ contribution based on the distance between the local gradient and global gradient. Specifically, we calculate the gradient distance for worker i as follows: M bi = Dis( ˜G,Gi) = j=1 Dis( ˜дj ,дj i ), (13) where Dis() is square of the Euclidean norm || ˜G,Gi||2”, wherein the Euclidean norm is a loss and bi is a marginal loss).
Regarding claim 6, the rejection of claims 1, 2, and 3 are incorporated and Gao further discloses wherein the participating node participating in the federated learning model comprises a plurality of participating nodes (Figures 1 and 3; the figures disclose a plurality of participating nodes or workers).
Regarding claim 7, the rejection of claims 1, 2, 3 and 6 are incorporated and Gao further discloses determining, based on a marginal loss of each participating node in the plurality of participating nodes and a normalization algorithm, the node mass of the participating node (§4.2; “we calculate workers’ contribution based on the distance between the local gradient and global gradient. Specifically, we calculate the gradient distance for worker i as follows: M bi = Dis( ˜G,Gi) = j=1 Dis( ˜дj ,дj i ), (13) where Dis() is square of the Euclidean norm || ˜G,Gi||2”, wherein the Euclidean norm is a loss and bi is a marginal loss; and §5.1; “and ωi(t) N (18) j=1 ωj(t) is the normalized weight of worker i’s rewards in t-th communication iteration”).
Regarding claim 8, the rejection of claims 1, 2, and 3 are incorporated and Gao further discloses wherein the participating node participating in the federated learning model comprises a plurality of participating nodes (Figures 1 and 3; the figures disclose a plurality of participating nodes or workers)
determining a participating node with a marginal loss greater than or equal to a predetermined loss threshold in the plurality of participating nodes as an effective participating node (§4.2; “we calculate workers’ contribution based on the distance between the local gradient and global gradient. Specifically, we calculate the gradient distance for worker i as follows: M bi = Dis( ˜G,Gi) = j=1 Dis( ˜дj ,дj i ), (13) where Dis() is square of the Euclidean norm || ˜G,Gi||2” and §4.3; “AssumingG0 is the gradient in which all values are 0, hence hav ing no utility to the system, to calculate the relative contribution, we set a threshold bh = Dis( ˜G,G0) to distinguish the positive contribution from negative contribution”).
Regarding claim 9, the rejection of claims 1, 2, 3, and 8 are incorporated and Gao further discloses performing, based on data volume information of the participating node and node mass of the participating node and to obtain the global gradient, an aggregation operation on a node local gradient of each of the effective participating node (§3.1; “The server aggregates local gradients to get the global gradient as follows: N ˜ G = where ni N i=1 ( ni N j=1 nj Gi), j=1 nj is the weight for worker i”).
Regarding claim 10, the rejection of claim 1 is incorporated and Gao further discloses wherein the participating node participating in the federated learning model comprises a plurality of participating nodes (Figures 1 and 3; the figures disclose a plurality of participating nodes or workers).
Regarding claim 11, the rejection of claims 1 and 10 are incorporated and Gao further discloses determining a node contribution degree of each of the plurality of participating nodes based on the node local gradient of the participating node and the global gradient; and determining a relative contribution degree of the participating node based on a node contribution degree of the participating node and the node contribution degree of each participating node (§3.1; and §4.3).
Regarding claim 12, the rejection of claims 1, 10, and 11 are incorporated and Gao further discloses obtaining a trustworthiness parameter of the participating node, wherein the trustworthiness parameter represents a comprehensive degree of reliability of the participating node in the target quantity of rounds of iterative calculations of the global gradient; and determining a reputation degree of the participating node based on the relative contribution degree and the trustworthiness parameter (§3.1; and §4.2).
Regarding claim 13, the rejection of claims 1, 10, 11, and 12 are incorporated and Gao further discloses determining a first quantity of times that the participating node is determined as an effective participating node in the target quantity of rounds of iterative calculations, wherein the effective participating node represents a participating node with a marginal loss greater than or equal to a predetermined loss threshold; determining a second quantity of times that the participating node is determined as an ineffective participating node in the target quantity of rounds of iterative calculations, wherein the ineffective participating node represents a participating node with a marginal loss less than the predetermined loss threshold; and determining the trustworthiness parameter of the participating node based on the first quantity of times and the second quantity of times (§3.1; and §4.2; and §4.3).
Regarding claim 14, the rejection of claims 1, 10, 11, and 12 are incorporated and Gao further discloses wherein the global gradient comprises a plurality of global gradient factors, wherein the participating node participating in the federated learning model comprises a plurality of participating nodes, and wherein the global gradient comprises the gradient obtained through the target quantity of rounds of iterative calculations (§3.1; “The server aggregates local gradients to get the global gradient as follows: N ˜ G = where ni N i=1 ( ni N j=1 nj Gi), j=1 nj is the weight for worker i. In iteration t, the parame ters of global model F is updated as follows: θt+1 = θt −η ˜Gt, where η is the learning rate, θt is the parameter in communication iteration t and ˜Gt is the global gradient. In the next iteration, workers train model from θt+1. The above steps iterate till the model converges”).
Regarding claim 15, the rejection of claims 1, 10, 11, 12, and 14 are incorporated and Gao further discloses determining a quantity of matching gradients corresponding to the participating node based on a ratio of the reputation degree of the participating node to a greatest reputation degree, wherein the greatest reputation degree represents the greatest reputation degree in reputation degrees of the plurality of participating nodes; and selecting global gradient factors of the quantity of matching gradients from the global gradient to obtain the actual model gradient (§3.1; and §4.2; and §4.3).
Regarding claim 16, the rejection of claims 1, 10, 11, 12, 14, and 15 are incorporated and Gao further discloses obtaining a node influence degree of each global gradient factor relative to the participating node, wherein the node influence degree of each global gradient factor indicates a degree that each global gradient factor is influenced by the participating node; and sorting, based on the node influence degree of each global gradient factor and to obtain sorted global gradient factors, the global gradient factors in the global gradient (§3.1; and §4.2; and §4.3; and Figures 11 and 12).
Regarding claim 17, the rejection of claims 1, 10, 11, 12, 14, 15, and 16 are incorporated and Gao further discloses selecting, based on a predetermined order and to obtain the actual model gradient, global gradient factors of the quantity of matching gradients from the sorted global gradient factors (§3.1; and §4.2; and §4.3; and Figures 11 and 12).
Regarding claim 18, the rejection of claims 1, 10, 11, 12, 14, 15, 16, and 17 are incorporated and Gao further discloses determining a first distribution parameter corresponding to each local gradient factor in the node local gradient of the participating node, wherein the first distribution parameter indicates a proportion of each local gradient factor in the node local gradient of the participating node; determining a second distribution parameter corresponding to each global gradient factor in the global gradient, wherein the second distribution parameter indicates a proportion of each global gradient factor in the global gradient; and, based on the first distribution parameter and the second distribution parameter, the node influence degree of each global gradient factor relative to the participating node (§3.1; and §4.2; and §4.3; and Figures 11 and 12).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
McMahan et al., “Communication-Efficient Learning of Deep Networks from Decentralized Data”, Feb. 28, 2017, arXiv:1602.05629v3, pp. 1-11.
Balakrishnan et al., (US 20230177349 A1).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Brent Hoover whose telephone number is (303)297-4403. The examiner can normally be reached Monday - Friday 9-5 MST.
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, Abdullah Kawsar can be reached at 571-270-3169. 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.
/BRENT JOHNSTON HOOVER/ Primary Examiner, Art Unit 2127
Prosecution Insights