DETAILED ACTION
Claims 1, 3-12, and 14 are presented for examination.
This office action is in response to submission of application on 17-NOVEMBER-2022.
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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 17-NOVEMBER-2022 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
The information disclosure statement (IDS) submitted on 19-SEPTEMBER-2025 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Amendment
The amendment filed 30-DECEMBER-2025 in response to the non-final office action mailed 30-SEPTEMBER-2025 has been entered. Claims 1, 3-12, and 14 remain pending in the application.
With regards to the non-final office action’s interpretation of claims under 112(f) and subsequent rejections under 112(b), the amendments to the claims have overcome the original rejection and render the claims no longer interpretable under 112(f).
With regards to the non-final office action’s rejection under 101, the amendments to the claims are not sufficient to overcome the original rejection with regards to the claims being directed towards an abstract idea.
With regards to the non-final office action’s rejection under 103, the amendment to the claims have overcome the original rejection. However, upon a new search for the amended limitations, a new 103 rejection over Qian in view of Satheesh further in view of Chen has been written. In light of the new art, the arguments are moot.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 5, 6, 8-10, and 12 rejected under 35 U.S.C. 101 because the claimed invention is direction to an abstract idea without significantly more.
MPEP 2106.04(a)(2)(Ill) “Accordingly, the "mental processes" abstract idea grouping is defined as concepts performed in the human mind, and examples of mental processes include observations, evaluations, Judgments, and opinions.
Further, the MPEP recites “The courts do not distinguish between mental processes that are performed entirely in the human mind and mental processes that require a human to use a physical aid (e.g., pen and paper or a slide run) to perform the claim limitation.
Regarding claim 5, which depends upon claim 1:
While claims 1 and 4 are not rejected as an abstract idea, its limitation are herein analyzed as the examiner believes that in light of the judicial exception believed to be present in claim 5, the limitations of claims 1 and 4 would be insignificant extra-solution activity.
Step 2A, Prong 1 will now be evaluated for this claim:
A judicial exception is recited in this claim as it recites a mathematical concept:
wherein at least one processor of the first information processing apparatus is configured to update a parameter of the third partial model based on the error information and backpropagation.
Backpropagation refers to a specific series of mathematical calculations, and the application of it here to update the parameter renders the updating of the parameters likewise a mathematical calculation.
Step 2A, Prong 2 will now be evaluated for this claim:
An information processing system, comprising- a first information processing apparatus and a second information processing apparatus configured to communicate with the first information processing apparatus via a network, wherein the information processing system being configured to perform learning processing on an inference model based on a neural network including an input layer, a plurality of intermediate layers, and an output layer, wherein the first information processing apparatus comprising: at least one processor; and at least one memory storing instructions which cause the processor to: acquire training data including learning data and a correct label; perform first learning processing by inputting the learning data to a first partial model including the input layer and a part of the plurality of intermediate layers of the inference model; and perform third learning processing on a third partial model including the output layer using an output obtained through second learning processing performed by the second information processing apparatus and the correct label, wherein the second information processing apparatus comprising; at least one processor; and at least one memory storing instructions which cause the processor to: perform the second learning processing by inputting an output obtained through the first learning processing to a second partial model including an intermediate layer that is included in the inference model and is different from the part of the plurality of intermediate layers included in the first partial model, wherein the first information processing apparatus is managed by a provider of the inference model, and the second information processing apparatus is managed by a user of the inference model, and wherein the first partial model and the third partial model are networks kept confidential from the user, and the second partial model is a network disclosed to the user.
Furthermore, the additional elements:
Claim 1: a first information processing apparatus and a second information processing apparatus configured to communicate with the first information processing apparatus via a network: the information processing apparatuses are both taken to be generic computers, supported by the list of generic computer components they comprise.
Claim 1: the information processing system being configured to perform learning processing on an inference model based on a neural network including an input layer, a plurality of intermediate layers, and an output layer: this limitation describes a generic neural network.
Claim 1: wherein the first information processing apparatus comprising: at least one processor; and at least one memory storing instructions which cause the processor to generic computer components.
Claim 1: perform first learning processing by inputting the learning data to a first partial model including the input layer and a part of the plurality of intermediate layers of the inference model – performing learning processing without further limitation is considered to be a generic computer function
Claim 1: perform third learning processing on a third partial model including the output layer using an output obtained through second learning processing performed by the second information processing apparatus and the correct label - performing learning processing without further limitation is considered to be a generic computer function.
Claim 1: the second information processing apparatus comprising: at least one processor, and at least one memory storing instruction which cause the processor to: is considered generic computer components.
perform the second learning processing by inputting an output obtained through the first learning processing to a second partial model including an intermediate layer that is included in the inference model and is different from the part of the plurality of intermediate layers included in the first partial model - performing learning processing without further limitation is considered to be a generic computer function.
wherein the first information processing apparatus is managed by a provider of the inference model, and the second information processing apparatus is managed by a user of the inference model – the management of information processing apparatus by particular entities does not make them a particular machine.
are interpreted as a general purpose computer under MPEP 2106.05(f)
Furthermore, MPEP 2106.05(g) Insignificant Extra-Solution Activity has found mere data gathering and post-solution activity to be insignificant extra-solution activity.
The following steps are mere data gathering:
Claim 1: acquire training data including learning data and a correct label – acquiring data is a form of data gathering.
The following steps are mere post-solution activity:
Claim 1: wherein the first partial model and the third partial model are networks kept confidential from the user, and the second partial model is a network disclosed to the user would be post-solution activity as it functionally demonstrate showing or sending the second partial model to the user, which would be a form of outputting.
The additional elements have been considered both individually and as an ordered combination in order to determine whether they integrate the exception into a practical application. Therefore, no meaningful limits are imposed practicing the abstract idea.
Therefore, the claim is related to an abstract idea.
Step 2B will now be discussed with regards to this claim:
The claim does not provide an inventive concept. There is no additional Insignificant Extra- Solution Activity, as identified in Step 2A Prong Two, that provides an inventive concept.
Adding insignificant extra-solution activity to the judicial exception, e.g., mere data gathering in conjunction with a law of nature or abstract idea such as a step of obtaining information about credit card transactions so that the information can be analyzed by an abstract mental process, as discussed in CyberSource v. Retail Decisions, Inc., 654 F.3d 1366, 1375, 99 USPQ2d 1690, 1694 (Fed. Cir. 2011) (see MPEP § 2106.05(g)) does not overcome a rejection.
The additional elements have been considered both individually and as an ordered combination as to whether they whether they warrant significantly more consideration.
The claim is ineligible.
Regarding claim 6, which depends upon claim 5:
The following would be mathematical calculation:
wherein the at least one processor of the second information processing apparatus is configured to update a parameter of the second partial model based on the error information transmitted from the first information processing apparatus, and wherein the at least one processor of the first information processing apparatus is configured to update a parameter of the first partial model based on the error information transmitted from the second information processing apparatus
As claim 6 depends upon claim 5, the updating described likewise results from backpropagation, which is considered to be a mathematical calculation.
This claim is ineligible.
Regarding claim 8, which depends upon claim 1:
This claim incorporates the insignificant extra-solution activity of claim 1. Furthermore, this claim is considered to be a mental process as the determination unit configured to determine if the learning is within an adequate range is accomplishable by the human mind, as the human mind is capable of comparing a result against a threshold and determining if the threshold is overcome.
This claim is ineligible.
Regarding claim 9, which depends upon claim 8:
This claim incorporates the insignificant extra-solution activity of claim 1. Furthermore, this claim is considered to be a mental process as the determination unit configured to determine if the learning is within an adequate range is accomplishable by the human mind, as the human mind is capable of comparing a result against a threshold and determining if the threshold is overcome.
This claim is ineligible.
Regarding claim 10, which depends upon claim 8:
This claim incorporates the insignificant extra-solution activity of claim 1. Furthermore, this claim is considered to be a mental process as the determination unit configured to determine if the learning is within an adequate range is accomplishable by the human mind, as the human mind is capable of comparing a result against a threshold and determining if the threshold is overcome.
This claim is ineligible.
Regarding claim 12, which depends upon claim 11:
This claim incorporates the insignificant extra-solution activity of claim 1. Furthermore, this claim is considered to be a mental process as the determination unit configured to determine if the learning is within an adequate range is accomplishable by the human mind, as the human mind is capable of comparing a result against a threshold and determining if the threshold is overcome.
Furthermore, as this claim depends upon claim 11, further consideration must be given to the limitation of the parent claim, which is not believed to contain a judicial exception. However, regarding the limitations of this claim:
An information processing system configured to make inference using the inference model based on the neural network that has been trained through the learning processing according to claim 1 and wherein the information processing system makes the inference on the data serving as the inference target using the first partial model, the second partial model, and the third partial model refers to the generic computer function of applying a machine learning model, which would be an instruction to “apply it".
wherein the at least one processor of the first information processing apparatus is further configured to acquire data serving as an inference target refers to the acquisition of data, which is data gathering.
As such in light of dependent claim 12, claim 11 is considered to be insignificant extra-solution activity.
This claim is ineligible.
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 1-8 and 11-14 are rejected under 35 U.S.C. 103 as being unpatentable over Qian et al. (Pub. No. US 20210374605 A1, filed October 30th 2020, hereinafter Qian) in view of Satheesh et al. (Pub. No. WO 2021064737 A1, filed October 4th 2019, hereinafter Satheesh) further in view of Chen et al. (Pub. No. US 20220092407 A1, filed September 23rd 2020, hereinafter Chen).
Regarding claim 1:
Claim 1 recites:
An information processing system, comprising- a first information processing apparatus and a second information processing apparatus configured to communicate with the first information processing apparatus via a network, wherein the information processing system being configured to perform learning processing on an inference model based on a neural network including an input layer, a plurality of intermediate layers, and an output layer, wherein the first information processing apparatus comprising: at least one processor; and at least one memory storing instructions which cause the processor to: acquire training data including learning data and a correct label; perform first learning processing by inputting the learning data to a first partial model including the input layer and a part of the plurality of intermediate layers of the inference model; and perform third learning processing on a third partial model including the output layer using an output obtained through second learning processing performed by the second information processing apparatus and the correct label, wherein the second information processing apparatus comprising; at least one processor; and at least one memory storing instructions which cause the processor to: perform the second learning processing by inputting an output obtained through the first learning processing to a second partial model including an intermediate layer that is included in the inference model and is different from the part of the plurality of intermediate layers included in the first partial model, wherein the first information processing apparatus is managed by a provider of the inference model, and the second information processing apparatus is managed by a user of the inference model, and wherein the first partial model and the third partial model are networks kept confidential from the user, and the second partial model is a network disclosed to the user.
Regarding the limitation an information processing system, comprising a first information processing apparatus and a second information processing apparatus configured to communicate with the first information processing apparatus via a network:
Qian teaches the use of federated learning, wherein a series of client devices with local models communicate with a remote server and remote shared model via a network (Paragraph 27). Here, the series of client devices would be the first information processing apparatus and the remote server the second information processing apparatus as they are two separate apparatuses that are in communication with each other via a network.
Regarding the limitation wherein the information processing system being configured to perform learning processing on an inference model based on a neural network including an input layer, a plurality of intermediate layers, and an output layer:
Qian teaches the use of a neural network, which would be a sort of inference model, that includes a number of LSTM layers that would comprise a plurality of intermediate layers, wherein a neural network in and of itself contains an input layer and output layer (Paragraph 24).
Regarding the limitation wherein the first information processing apparatus comprising: at least one processor; and at least one memory storing instructions which cause the processor to: acquire training data including learning data and a correct label:
Qian teaches supervised learning algorithms that use labeled examples as training data in order to train a machine learning model (Paragraph 106). The labeled examples would comprise learning data and a correct label and are therefore the acquired training data.
Regarding the limitation perform first learning processing by inputting the learning data to a first partial model including the input layer and a part of the plurality of intermediate layers of the inference model:
Qian teaches federated learning, wherein in the first information processing apparatus (the client devices) an individual client device may train a machine learning model based on its own user data (i.e., inputting learning user data to a first model through the input layer) wherein the model learns gradients for the client device’s data through a plurality of intermediate layers (Paragraph 27). This model would be considered a partial model as it only has access to partial data, and as such does not produce the full inference result.
Regarding the limitation perform third learning processing on a third partial model including the output layer using an output obtained through second learning processing performed by the second information processing apparatus and the correct label:
Qian teaches federated learning, wherein the first model discussed above may send its gradients to the second information processing apparatus, wherein a second model (the remote shared model) produces an output through second learning that is then sent back to the client devices (Paragraph 27). Therefore, another client device now has learning from the first client device. This new client device would contain its own third partial model that produces gradients in the same manner as the first partial model in order to continue the learning process as an output from an output layer.
Qian discloses wherein the second information processing apparatus comprising; at least one processor; and at least one memory storing instructions which cause the processor to: perform the second learning processing by inputting an output obtained through the first learning processing to a second partial model:
Qian teaches federated learning, wherein the first model discussed above may send its gradients to the second information processing apparatus, wherein a second model (the remote shared model) on the remote server (or the second learning unit) performs a second learning process using the gradients that were sent from the first partial model as an output (Paragraph 27). However, Qian does not teach that the intermediate layers of the second model are different than the intermediate layers of the first model.
Qian discloses wherein the first information processing apparatus is managed by a provider of the inference model, and the second information processing apparatus is managed by a user of the inference model:
Qian teaches the use of federated learning for keeping user data on-device (Paragraph 27) which indicates the differentiation between the global model managed by a provider of the federated learning models, and the a secondary device where the data is kept is managed by a user.
Satheesh discloses including an intermediate layer that is included in the inference model and is different from the part of the plurality of intermediate layers included in the first partial model:
Satheesh in the same field of endeavor of distributed machine learning teaches two models with different sets of layers (Paragraph 10), and therefore contain an intermediate layer that is included in the inference model and is different from the part of the plurality of intermediate layers included in the first partial model.
Satheesh and the present application are analogous art because they are in the same field of endeavor of distributed machine learning.
Chen discloses wherein the first partial model and the third partial model are networks kept confidential from the user, and the second partial model is a network disclosed to the user:
Chen in the same field of endeavor of machine learning teaches that a machine learning model may be an access-limited machine learning model where users to not have full access to its architecture (Paragraph 52). This demonstrates a machine learning model wherein some parts of the model -– e.g., the first partial model and third martial model – are kept confidential from the user, but some parts – e.g., the second partial model – are disclosed to the user.
Chen and the present application are analogous art because they are in the same field of endeavor of distributed machine learning.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to implement a system that implemented the teachings of Qian, the teachings of Satheesh, and the teachings of Chen. This would have provided the advantage of optimization within a federated learning framework (Satheesh, Paragraph 6) as well as allowing flexibility in whether or not to provide users access to certain model data (Chen, Paragraph 66).
Regarding claim 3, which depends upon claim 1:
Claim 3 recites:
The information processing system according to claim 1, wherein the at least one processor of the first information processing apparatus is configured to acquire the training data from the second information processing apparatus.
Qian in view of Satheesh further in view of Chen discloses the system of claim 1 upon which claim 3 depends. Furthermore, Satheesh discloses the limitation of claim 3:
Satheesh teaches a federated learning system wherein the global model, analogous to the remote shared model of Qian, receives equations for training the model (Paragraph 46) wherein the equations act as a form of training data acquired by the second information processing apparatus, as the global / remote shared model acts as the second information processing apparatus as seen in Qian’s disclosure of claim 1.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to implement a system that implemented the teachings of Qian, the teachings of Satheesh, and the teachings of Chen. This would have provided the advantage of optimization within a federated learning framework (Satheesh, Paragraph 6) as well as allowing flexibility in whether or not to provide users access to certain model data (Chen, Paragraph 66).
Regarding claim 4, which depends upon claim 1:
Claim 4 recites:
The information processing system according to claim 1, wherein the at least one processor of the first information processing apparatus is configured to acquire the error information based on the correct label and an output from the output layer
Qian in view of Satheesh further in view of Chen discloses the system of claim 1 upon which claim 4 depends. Furthermore, Qian discloses the limitation of claim 4:
Qian teaches supervised learning algorithms which acquire error information by comparing an acquired correct label from the training data to the output from the output layer (Paragraph 106) wherein the supervised learning algorithm may be used in the third learning unit.
Regarding claim 5, which depends upon claim 4:
Claim 5 recites:
The information processing system according to claim 4, wherein at least one processor of the first information processing apparatus is configured to update a parameter of the third partial model based on the error information and backpropagation.
Qian in view of Satheesh further in view of Chen discloses the system of claim 4 upon which claim 5 depends. Furthermore, Satheesh discloses the limitation of claim 5:
Satheesh teaches that the user devices’ models of its federated learning system use a backpropagation technique, which in and of itself uses the error information (Paragraph 58). As user device models have previously been used in Qian to provide a third learning unit, the at least one processor of the first information processing apparatus may in combination update a parameter in this manner.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to implement a system that implemented the teachings of Qian, the teachings of Satheesh, and the teachings of Chen. This would have provided the advantage of optimization within a federated learning framework (Satheesh, Paragraph 6) as well as allowing flexibility in whether or not to provide users access to certain model data (Chen, Paragraph 66).
Regarding claim 6, which depends upon claim 5:
Claim 6 recites:
The information processing system according to wherein the at least one processor of the second information processing apparatus is configured to update a parameter of the second partial model based on the error information transmitted from the first information processing apparatus, and wherein the at least one processor of the first information processing apparatus is configured to update a parameter of the first partial model based on the error information transmitted from the second information processing apparatus.
Qian in view of Satheesh further in view of Chen discloses the system of claim 5 upon which claim 6 depends. Furthermore, Qian discloses wherein the at least one processor of the second information processing apparatus is configured to update a parameter of the second partial model based on the error information transmitted from the first information processing apparatus:
Qian teaches that remote server, or the at least one processor of the second information processing apparatus, aggregates gradient and weight information from each client system, which would include the first information processing apparatus’s transmitted information. Furthermore, this information is then used to update the new central model, which would updating a parameter of the second partial model based on the error information (Paragraph 35).
Qian discloses wherein the at least one processor of the first information processing apparatus is configured to update a parameter of the first partial model based on the error information transmitted from the second information processing apparatus:
Qian teaches that this updated central model may then be transmitted to the client devices, which would include the first partial model, wherein the distribution of the model would update the parameters of the first partial model as it is received and incorporated in the client models (Paragraph 35).
Regarding claim 7, which depends upon claim 1:
Claim 7 recites:
The information processing system according to claim 1, wherein the at least one processor of the second information processing apparatus in the second information processing apparatus is configured to generate the second partial model by performing additional learning with parameters of the first partial model and the third partial model being fixed
Qian in view of Satheesh further in view of Chen discloses the system of claim 1 upon which claim 7 depends. Furthermore, Qian discloses the limitation of claim 7:
Qian teaches that the remote server model, the second learning unit in the second information processing apparatus, aggregates information from the client system and updates the model, generating the second partial model by performing additional learning (Paragraph 35). The parameters of the first partial model and the third partial model are fixed as they are not simultaneously updated with the shared model.
Regarding claim 8, which depends upon claim 1:
Claim 8 recites:
The information processing system according to claim 1, further comprising an information processing system is configured to determine whether the first information processing apparatus performs learning, in an adequate range, for at least either the training data or a partial model
Qian in view of Satheesh further in view of Chen discloses the system of claim 1 upon which claim 8 depends. Furthermore, Qian discloses the limitation of claim 8:
Qian teaches that training may be continued depending on the performance of the new central model (Paragraph 35), which demonstrates a determination of when learning is within an adequate range for the partial model.
Regarding claim 11, which depends upon claim 1:
Claim 11 recites:
An information processing system configured to make inference using the inference model based on the neural network that has been trained through the learning processing according to claim 1, wherein the at least one processor of the first information processing apparatus is further configured to acquire data serving as an inference target, and wherein the information processing system makes the inference on the data serving as the inference target using the first partial model, the second partial model, and the third partial model
Qian in view of Satheesh further in view of Chen discloses the system of claim 1 upon which claim 11 depends. Furthermore, Qian discloses make inference using the inference model based on the neural network that has been trained through the learning processing according to claim 1, wherein the at least one processor of the first information processing apparatus is further configured to acquire data serving as an inference target:
Qian teaches that using the above model or the inference model based on the neural network that has been trained through the learning processing according to claim 1, predictions may be made about outputs values, which would be an inference (Paragraph 106). This may further be done with a training dataset that is used by the client models, or the first information processing apparatus (Paragraph 106), that is a supervised learning training dataset wherein there is a known inference target.
Qian discloses wherein the information processing system makes the inference on the data serving as the inference target using the first partial model, the second partial model, and the third partial model:
Qian teaches that using the above model or the inference model based on the neural network that has been trained through the learning processing according to claim 1, predictions may be made about outputs values, which would be an inference (Paragraph 106) wherein the process of claim 1 describes the use of the first, second, and third partial model.
Regarding claim 12, which depends upon claim 11:
Claim 12 recites:
The information processing system according to claim 11, wherein the at least one processor of the first information processing apparatus is configured to determine whether the inference is made, in an adequate range, for at least one of the data as the inference target or an inference result
Qian in view of Satheesh further in view of Chen discloses the system of claim 11 upon which claim 12 depends. Furthermore, Qian discloses the limitation of claim 12:
Qian teaches that training may be continued depending on the performance of the new central model (Paragraph 35), which demonstrates a determination of when learning is within an adequate range for the partial model.
Claim 14 recites a method that parallels the system of claim 1. Therefore, the analysis discussed above with respect to claim 1 also applies to claim 14. Accordingly, claim 14 is rejected based on substantially the same rationale as set forth above with respect to claim 1.
Claims 9 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Qian in view of Satheesh further in view of Chen, further in view of Ronen et al. (Pub. No. US 20210235287 A1, filed April 14th 2021, hereinafter Ronen)
Regarding claim 9, which depends upon claim 8:
Claim 9 recites:
The information processing system according to claim 8, wherein the information processing system is configured to determine whether the first information processing apparatus performs the learning, in the adequate range, for at least either the training data or a partial model, depending on whether a component ratio of the correct label included in the training data satisfies a predetermined criterion
Qian in view of Satheesh further in view of Chen discloses the system of claim 8 upon which claim 9 depends. Furthermore:
Qian in view of Satheesh further in view of Chen has previously taught wherein the information processing system is configured to determine whether the first information processing apparatus performs the learning, in the adequate range, for at least either the training data or a partial model in claim 8.
However, Ronen discloses depending on whether a component ratio of the correct label included in the training data satisfies a predetermined criterion:
Ronen in the same field of endeavor of machine learning teaches that for a series of data belonging to the same class, i.e. with the same correct label, performance may be evaluated to ensure that it is above a particular threshold or predetermined criterion (Paragraph 62). Performance would be analogous to a component ratio as the performance is used to identify anomalies (Paragraph 62) as the component ratio is in the present application’s specification, wherein it is used to avoid over- or underfitting the model (present application, paragraph 42).
Ronen and the present application are analogous art because they are in the same field of endeavor, machine learning.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to implement a system that implemented the teachings of Qian, the teachings of Satheesh, the teachings of Chen, and the teachings of Ronen. This would have provided the advantage of optimization within a federated learning framework (Satheesh, Paragraph 6), allowing flexibility in whether or not to provide users access to certain model data (Chen, Paragraph 66), and further customization and control of performance metrics (Ronen, Paragraph 14).
Regarding claim 10, which depends upon claim 8:
Claim 10 recites:
The information processing system according to claim 8, the information processing system is configured to determine whether the first information processing apparatus performs the learning, in the adequate range, for at least either the training data or a partial model, depending on whether a variation in parameter of the partial model due to the learning satisfies a predetermined criterion
Qian in view of Satheesh further in view of Chen discloses the system of claim 8 upon which claim 10 depends. Furthermore:
Qian in view of Satheesh has previously taught wherein the information processing system is configured to determine whether the first information processing apparatus performs the learning, in the adequate range, for at least either the training data or a partial model in claim 8.
However, Ronen discloses depending on whether a variation in parameter of the partial model due to the learning satisfies a predetermined criterion:
Ronen in the same field of endeavor of machine learning teaches that for a series of data belonging to the same class, i.e. with the same correct label, performance may be evaluated to ensure that it is above a particular variation threshold or predetermined criterion (Paragraph 62), wherein the variation threshold detects anomalies and hence would measure variation of a parameter of the partial model.
Ronen and the present application are analogous art because they are in the same field of endeavor, machine learning.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to implement a system that implemented the teachings of Qian, the teachings of Satheesh, the teachings of Chen, and the teachings of Ronen. This would have provided the advantage of optimization within a federated learning framework (Satheesh, Paragraph 6), allowing flexibility in whether or not to provide users access to certain model data (Chen, Paragraph 66), and further customization and control of performance metrics (Ronen, Paragraph 14).
Response to Arguments
Applicant’s arguments filed 30-DECEMBER-2025 have been fully considered, but the examiner believes that not all are fully persuasive.
Regarding the applicant’s remarks on non-final office action’s 101 rejection of the claims as an abstract idea, the applicant argues that regardless of the original claims, the amended claims are not directed towards an abstract idea. The examiner respectfully requests the applicant’s consideration of the following:
The applicant argues that claims 5, 6, 8-10, and 12 are allowable and analyzes claim 1 from which these claims depend under 101. The examiner will address these arguments:
The applicant states that “The claims do not recite a judicial exception under Step 2A, Prong One” and “Claim 1 recites a concrete technical arrangement” that is not a mathematical concept, method of organizing human activity, or a mental process. The examiner agrees the applicant, as claim 1 is not rejected as a judicial exception. However, the rejection of the dependent claims is not traversed.
The applicant further argues that “even if a judicial exception were considered recited, the claims integrate it into a practical application”. However, while the applicant state that “these operations [of claim 1] improve the functioning of the device or system itself” no improvement is cited in the arguments, shown in the claims, or discussed in the specification. Therefore, the examiner does not consider the abstract ideas of the independent claims to be integrated into a practical application by virtue of an improvement to the technology in claim 1.
The applicant argues that “the claims include significantly more” and cites the claim as a whole, but does not argue which aspects of the claim may be beyond insignificant extra-solution activity when considered in light of the abstract ideas believed to be present in the rejected dependent claims. In the absence of specific arguments, the examiner respectfully requests that the applicant reference the previous 101 rejection included within, which addresses why the limitations and the claim language of claim 1 are not considered to amount to significantly more.
The applicant further states “USPTO ‘Close Call’ Standard Requires Withdrawal of the Rejection’. The examiner believes that it is ‘more likely than not that the claim(s) are illegible’ and therefore believes the 101 rejections to be proper, even in light of this guidance.
Regarding the applicant’s remarks on the non-final office action’s 103 rejection of the claims, the applicant argues that Qian in view of Satheesh do not teach the amended limitations of these claims. As such, the applicant argues that all claims dependent on the above would additionally not be obvious under 103. The examiner agrees that the prior art of the original office action does not teach the amended limitations. However, upon a new search of the prior art for the amended limitations, the examiner has written a new rejection under 103 to address these limitations and respectfully requests applicant’s consideration of the following:
The applicant argues that Qian “does not distinguish management entities such as a model provider and model users” and Satheesh “does not disclose a structural division in which a ‘second partial model including an intermediate layer … different from the part … included in the first model’ is arranged only on a user apparatus, while partial models including the input layer and the output layer are retained and executed only on a provider-side apparatus” in reference to the newly amended limitations. The newly amended limitations are addressed below by Chen.
Previously disclosed art Qian discloses wherein the first information processing apparatus is managed by a provider of the inference model, and the second information processing apparatus is managed by a user of the inference model:
Qian teaches the use of federated learning for keeping user data on-device (Paragraph 27) which indicates the differentiation between the global model managed by a provider of the federated learning models, and the a secondary device where the data is kept is managed by a user.
New art Chen discloses wherein the first partial model and the third partial model are networks kept confidential from the user, and the second partial model is a network disclosed to the user:
Chen in the same field of endeavor of machine learning teaches that a machine learning model may be an access-limited machine learning model where users to not have full access to its architecture (Paragraph 52). This demonstrates a machine learning model wherein some parts of the model -– e.g., the first partial model and third martial model – are kept confidential from the user, but some parts – e.g., the second partial model – are disclosed to the user.
Chen and the present application are analogous art because they are in the same field of endeavor of distributed machine learning.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALEXANDRIA JOSEPHINE MILLER whose telephone number is (703)756-5684. The examiner can normally be reached Monday-Thursday: 7:30 - 5:00 pm, every other Friday 7:30 - 4:00.
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, Mariela Reyes can be reached at (571) 270-1006. 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.
/A.J.M./Examiner, Art Unit 2142 /Mariela Reyes/Supervisory Patent Examiner, Art Unit 2142
Prosecution Insights