DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This action is responsive to the application filed on March 13th, 2024. Claims 1-15 are pending in the case. Claims 1, 13, and 15 are independent claims.
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The listing of references in the specification, page 19 lines 19-29, is not a proper information disclosure statement. 37 CFR 1.98(b) requires a list of all patents, publications, or other information submitted for consideration by the Office, and MPEP § 609.04(a) states, "the list may not be incorporated into the specification but must be submitted in a separate paper." Therefore, unless the references have been cited by the examiner on form PTO-892, they have not been considered.
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 7 and 8 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 7 recites the limitation "indicating the attribute of the model". There is insufficient antecedent basis for this limitation in the claim. For examination purposes, this limitation has been interpreted as “indicating an
Claim 8 is rejected for being dependent on a rejected base claim without curing any of the deficiencies.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding claim 1:
Step 1: Claim 1 is directed to [a]n information processing apparatus, therefore it falls under the statuary category of a manufacture.
Step 2A Prong 1: The claim recites, in part:
“determine whether at least one of a plurality of models satisfies the conditions” this encompasses the mental determination of whether any observed models satisfy observed conditions.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“processing circuitry configured to”, “the plurality of models being each generated by executing learning processing using learning data” the limitations are an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP § 2106.05(f)(2).
“receive, from a terminal device, condition information indicating conditions related to a model”, “transmit to the terminal device model specifying information that specifies a one of the models satisfying the conditions based on a result of the determination” these limitations are an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g).
Step 2B: The additional elements “processing circuitry configured to”, “the plurality of models being each generated by executing learning processing using learning data”, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. “receive, from a terminal device, condition information indicating conditions related to a model”, “transmit to the terminal device model specifying information that specifies a one of the models satisfying the conditions based on a result of the determination” these limitations are an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d. Therefore, the claim is ineligible.
Regarding claim 2, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“wherein the plurality of models include a plurality of global models to be updated based on a plurality of local models” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Therefore, the claim is ineligible.
Regarding claim 3, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“the plurality of models include a plurality of local models each generated by corresponding one of a plurality of nodes” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Therefore, the claim is ineligible.
Regarding claim 4, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: The claim recites, in part:
“wherein the conditions include a priority of an attribute of the model” a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more.
Regarding claim 5, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: The claim recites, in part:
“wherein the attribute is one of a fee for using the model and an evaluation of the model” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
Step 2A Prong 2: The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more.
Regarding claim 6, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: The claim recites, in part:
“wherein the conditions include a task executed by the model” a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more.
Regarding claim 7, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“the model specifying information includes model attribute information indicating the attribute of the model” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Therefore, the claim is ineligible.
Regarding claim 8, the rejection of claim 7 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“wherein the attribute of the model includes at least one of a number of the learning data used for generating the model and a business type of a client of a node that generates the model, the model being a local model” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Therefore, the claim is ineligible.
Regarding claim 9, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“the model specifying information includes service attribute information indicating an attribute of a service provider providing the model” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Therefore, the claim is ineligible.
Regarding claim 10, the rejection of claim 9 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“wherein the attribute of the service provider includes at least one of a number of local models and an availability of the local model used for updating a global model” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Therefore, the claim is ineligible.
Regarding claim 11, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“wherein the processing circuitry is configured to further receive model selection information indicating a model selected from among one or more models satisfying the conditions from the terminal device” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. “wherein the processing circuitry is configured to further receive model selection information indicating a model selected from among one or more models satisfying the conditions from the terminal device” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d. Therefore, the claim is ineligible.
Regarding claim 12, the rejection of claim 11 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“transmit the model selection information to a device of a service provider that provides the model” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g).
Step 2B: The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. “transmit the model selection information to a device of a service provider that provides the model” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d. Therefore, the claim is ineligible.
Regarding claim 13:
Step 1: Claim 13 is directed to [a]n information processing method, therefore it falls under the statuary category of a process.
Step 2A Prong 1: The claim recites, in part:
“determining whether at least one of a plurality of models satisfies the conditions” this encompasses the mental determination of whether any observed models satisfy observed conditions.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“the plurality of models being each generated by executing learning processing using learning data” the limitation is an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP § 2106.05(f)(2).
“receiving, from a terminal device, condition information indicating conditions related to a model”, “transmitting to the terminal device model specifying information that specifies a one of the models satisfying the conditions based on a result of the determination” these limitations are an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g).
Step 2B: The additional elements “the plurality of models being each generated by executing learning processing using learning data”, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. “receiving, from a terminal device, condition information indicating conditions related to a model”, “transmitting to the terminal device model specifying information that specifies a one of the models satisfying the conditions based on a result of the determination” these limitations are an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d. Therefore, the claim is ineligible.
Regarding claim 14, the rejection of claim 1 is incorporated and further:
Step 2A Prong 1: a continuation of the abstract idea identified in the parent claim.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“a terminal device” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h).
“terminal device processing circuitry configured to”, “cause a display to display the model specifying information received from the information processing apparatus” the limitations are an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP § 2106.05(f)(2).
“the information processing apparatus being configured to communicate with the terminal device”, “receive a setting operation for setting conditions related to the model”, “transmit condition information indicating the conditions set by the setting operation to the information processing apparatus” these limitations are an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g).
Step 2B: The additional elements “a terminal device”, “terminal device processing circuitry configured to”, “cause a display to display the model specifying information received from the information processing apparatus”, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. “the information processing apparatus being configured to communicate with the terminal device”, “receive a setting operation for setting conditions related to the model”, “transmit condition information indicating the conditions set by the setting operation to the information processing apparatus” these limitations are an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d. Therefore, the claim is ineligible.
Regarding claim 15:
Step 1: Claim 15 is directed to [a]n information processing system, therefore it falls under the statuary category of a machine.
Step 2A Prong 1: The claim recites, in part:
“determine whether at least one of a plurality of models satisfies the conditions set by the setting operation” this encompasses the mental determination of whether any observed models satisfy observed conditions.
Step 2A Prong 2: The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows:
“receive a setting operation for setting conditions related to a model” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g).
“the plurality of models being each generated by executing learning processing using learning data”, “cause a display to display model specifying information that specifies one of a plurality of models satisfying the conditions set by the setting operation” the limitations are an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP § 2106.05(f)(2).
Step 2B: The additional elements “the plurality of models being each generated by executing learning processing using learning data”, “cause a display to display model specifying information that specifies one of a plurality of models satisfying the conditions set by the setting operation”, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. “receive a setting operation for setting conditions related to a model” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d. Therefore, the 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 1-13 are rejected under 35 U.S.C. § 103 as being unpatentable over Deng et al. (“AUCTION: Automated and Quality-Aware Client Selection Framework for Efficient Federated Learning”, Deng et al., 24 Dec. 2021) (hereinafter “Deng “) in view of Appel et al. (US 2022/0188663 A1) (hereinafter “Appel”).
Regarding claim 1:
Deng teaches [a]n information processing apparatus comprising:
processing circuitry configured to: receive, from a terminal device, condition information indicating conditions related to a model (Deng, page 6, col 1, ¶1 “To facilitate high-quality client selection within a limited budget, following the principles in Section 3, we set the features xi of each client Ci as a 4-dimensional vector represented by
x
i
=
ⅆ
i
,
q
i
l
,
q
i
d
,
b
i
, where di are the number of data samples used for training,
q
i
l
and
q
i
d
are the data quality in terms of data labels and data distribution, respectively, and bi is the required price (i.e., payment) for client Ci to complete the learning task.” Here, the state received from each client relating to its local model can be considered the condition information indicating conditions related to a model);
determine whether at least one of a plurality of models satisfies the conditions (Deng, page 3, col 1, section 2.1, ¶1, “Therefore, to achieve satisfying FL performance, the platform needs to select an optimal subset of clients from C within the budget B for the given FL task. The selected clients can collaboratively train the FL model using their local data samples and then receive their declared payments.” Here, models on clients within the selected optimal subset can be considered the models that satisfy the conditions), the plurality of models being each generated by executing learning processing using learning data (Deng, page 3, col 2, ¶3 “In each round r, based on the global model
w
r
, each participating client conducts model training individually by using the local data set Di, the training results of which can be used to update the local model parameters
w
i
r
.”);
Deng does not teach “transmit to the terminal device model specifying information that specifies a one of the models satisfying the conditions based on a result of the determination”
However, Appel teaches transmit to the terminal device model specifying information that specifies a one of the models satisfying the conditions based on a result of the determination (Appel, ¶5 “program instructions to determine acceptable architectural templates to evaluate the machine learning models based on the input; program instructions to determine a first list of architectures and metrics based on a calculation of maximum neural network sizes of the acceptable architectural templates not exceeding the constraint; and program instructions to send the first list of architectures and metrics to the user for selection.” Here, the sending of the model architectures to the user based on a constraint can be considered the transmission of specifying information satisfying conditions based on the determination).
Deng and Appel are analogous art because both references concern methods for machine learning model selection based on conditions. Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Deng’s federated learning system to incorporate the selection and transmission of model specifying information taught by Appel. The motivation for doing so would have been to select models that fit a user’s needs, such as their constraints, as stated in Appel, ¶63 “… users do not know how long their machine learning models will take to run at inference time and, accordingly, the users do not know how much it will cost to run the selected machine learning models. Similarly, the time and associated cost to train the selected machine learning models is also treated as a black box. When machine learning models are deployed at scale, this is a problem, because the machine learning models can be too costly to execute, either because they take too long to run, or because they require large amounts of computational resources.” Furthermore, Appel, ¶18 “the present invention can provide a technique of performing an architecture search for machine learning models while respecting constraints provided by users, when combined with knowledge about performing the user's desired task.”
Regarding claim 2:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 1,
wherein the plurality of models include a plurality of global models to e updated based on a plurality of local models (Deng, page 7-8, col 2, section 5.2, ¶1 “Specifically, in each round, each selected client Ci trains the global model with di local data samples and commits the model updates to the FL platform for aggregation.”).
Regarding claim 3:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 1,
wherein the plurality of models include a plurality of local models each generated by corresponding one of a plurality of nodes (Deng, page 3, col 2, ¶3 “In each round r, based on the global model
w
r
, each participating client conducts model training individually by using the local data set Di, the training results of which can be used to update the local model parameters
w
i
r
.” Here, each client can be considered a node which generates a local model).
Regarding claim 4:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 1,
wherein the conditions include a priority of an attribute of the model (Deng, page 10, col 1, ¶1 “It means that the Greedy mechanism prioritizes clients with high data size and low price, while AUCTION emphasizes more importance on the data quality, which can achieve better performance.” Here, the prioritization of the data size and prices in the Greedy mechanism can be considered the prioritization of model attributes).
Regarding claim 5:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 4,
wherein the attribute is one of a fee for using the model and an evaluation of the model (Deng, page 6, col 1, ¶1 “To facilitate high-quality client selection within a limited budget, following the principles in Section 3, we set the features xi of each client Ci as a 4-dimensional vector represented by
x
i
=
ⅆ
i
,
q
i
l
,
q
i
d
,
b
i
, where di are the number of data samples used for training,
q
i
l
and
q
i
d
are the data quality in terms of data labels and data distribution, respectively, and bi is the required price (i.e., payment) for client Ci to complete the learning task.” Here the required price can be considered the fee and the data quality can be considered the evaluation of the model);.
Regarding claim 6:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 1,
wherein the conditions include a task executed by the model (Deng, page 5, col 2, 4.2, ¶2 “The state captures the status of candidate clients for a given FL task, including the data quality, data size, and the claimed price, while the action determines which clients are selected to participate in the task.” Here, the FL task can be considered the task).
Regarding claim 7:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 1,
wherein the model specifying information includes model attribute information indicating the attribute of the model (Deng, page 6, col 1, ¶1 “To facilitate high-quality client selection within a limited budget, following the principles in Section 3, we set the features xi of each client Ci as a 4-dimensional vector represented by
x
i
=
ⅆ
i
,
q
i
l
,
q
i
d
,
b
i
, where di are the number of data samples used for training,
q
i
l
and
q
i
d
are the data quality in terms of data labels and data distribution, respectively, and bi is the required price (i.e., payment) for client Ci to complete the learning task.” Here di- - the number of data samples used for training - can be considered an attribute of the model).
Regarding claim 8:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 7,
wherein the attribute of the model includes at least one of a number of the learning data used for generating the model (Deng, page 6, col 1, ¶1 “To facilitate high-quality client selection within a limited budget, following the principles in Section 3, we set the features xi of each client Ci as a 4-dimensional vector represented by
x
i
=
ⅆ
i
,
q
i
l
,
q
i
d
,
b
i
, where di are the number of data samples used for training,
q
i
l
and
q
i
d
are the data quality in terms of data labels and data distribution, respectively, and bi is the required price (i.e., payment) for client Ci to complete the learning task.” Here di- - the number of data samples used for training - can be considered a number of the learning data used for generating the model. It is noted the claim recites alternative language, and Deng teaches at least one of the alternatives.) and a business type of a client of a node that generates the model, the model being a local model.
Regarding claim 9:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 1,
wherein the model specifying information includes service attribute information indicating an attribute of a service provider providing the model (Deng, page 3, col 1, section 2.1, ¶1 “For a given FL task, there exist N clients denoted by
c
=
c
1
,
c
2
,
…
,
c
n
, willing to participate in the distributed learning with a claimed price
b
1
,
b
2
,
…
,
b
n
, and each client Ci has a set of private local data samples Di relevant to the FL task.” Here, the number of clients N can be considered an attribute of a service provider providing the model).
Regarding claim 10:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 9,
wherein the attribute of the service provider includes at least one of a number of local models (Deng, page 3, col 1, section 2.1, ¶1 “For a given FL task, there exist N clients denoted by
c
=
c
1
,
c
2
,
…
,
c
n
, willing to participate in the distributed learning with a claimed price
b
1
,
b
2
,
…
,
b
n
, and each client Ci has a set of private local data samples Di relevant to the FL task.” Here, the number of clients N can be considered a number of local models. It is noted the claim recites alternative language, and Deng teaches at least one of the alternatives.) and an availability of the local model used for updating a global model.
Regarding claim 11:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 1,
wherein the processing circuitry is configured to further receive model selection information indicating a model selected from among one or more models satisfying the conditions from the terminal device (Deng, page 3, col 1-2, section 2.1, ¶2-3, “A learning task is submitted to the FL platform, and there is a limited budget B which can be used to recruit clients to update the parameters w of the global learning model based on their local training results. Client initialization. The set of clients C that are willing to participate in the task, i.e., candidate clients, report their client-side information and prices, which will be used for the client selection in the next step.” Here, the clients, each of which has a local mode, reporting of the client-side information can be considered the receiving of model selection information indicating a model selected).
Regarding claim 12:
Deng in view of Appel teaches [t]he information processing apparatus according to claim 11,
wherein the processing circuitry is configured to transmit the model selection information to a device of a service provider that provides the model (Deng, page 3, col 2, ¶2 “Client selection. The FL platform conducts client selection to choose a subset of participants from the candidate clients, and then delivers the initial global model w0 to the selected participating clients.” Here, the delivery of the global model to clients selected can be considered the transmitting of model selection information, and the budgeted clients can be considered a service provider that provides the model).
Regarding claim 13:
Deng teaches [a]n information processing method, comprising:
receiving, from a terminal device, condition information indicating conditions related to a model (Deng, page 6, col 1, ¶1 “To facilitate high-quality client selection within a limited budget, following the principles in Section 3, we set the features xi of each client Ci as a 4-dimensional vector represented by
x
i
=
ⅆ
i
,
q
i
l
,
q
i
d
,
b
i
, where di are the number of data samples used for training,
q
i
l
and
q
i
d
are the data quality in terms of data labels and data distribution, respectively, and bi is the required price (i.e., payment) for client Ci to complete the learning task.” Here, the state received from each client relating to its local model can be considered the condition information indicating conditions related to a model);
determining whether at least one of a plurality of models satisfies the conditions (Deng, page 3, col 1, section 2.1, ¶1, “Therefore, to achieve satisfying FL performance, the platform needs to select an optimal subset of clients from C within the budget B for the given FL task. The selected clients can collaboratively train the FL model using their local data samples and then receive their declared payments.” Here, models on clients within the selected optimal subset can be considered the models that satisfy the conditions), the plurality of models being each generated by executing learning processing using learning data (Deng, page 3, col 2, ¶3 “In each round r, based on the global model
w
r
, each participating client conducts model training individually by using the local data set Di, the training results of which can be used to update the local model parameters
w
i
r
.”);
Deng does not teach “transmitting to the terminal device model specifying information that specifies one of the models satisfying the conditions based on a result of the determination”
However, Appel teaches transmit to the terminal device model specifying information that specifies a one of the models satisfying the conditions based on a result of the determination (Appel, ¶5 “program instructions to determine acceptable architectural templates to evaluate the machine learning models based on the input; program instructions to determine a first list of architectures and metrics based on a calculation of maximum neural network sizes of the acceptable architectural templates not exceeding the constraint; and program instructions to send the first list of architectures and metrics to the user for selection.” Here, the sending of the model architectures to the user based on a constraint can be considered the transmission of specifying information satisfying conditions based on the determination).
Deng and Appel are analogous art because both references concern methods for machine learning model selection based on conditions. Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Deng’s federated learning system to incorporate the selection and transmission of model specifying information taught by Appel. The motivation for doing so would have been to select models that fit a user’s needs, such as their constraints, as stated in Appel, ¶63 “… users do not know how long their machine learning models will take to run at inference time and, accordingly, the users do not know how much it will cost to run the selected machine learning models. Similarly, the time and associated cost to train the selected machine learning models is also treated as a black box. When machine learning models are deployed at scale, this is a problem, because the machine learning models can be too costly to execute, either because they take too long to run, or because they require large amounts of computational resources.” Furthermore, Appel, ¶18 “the present invention can provide a technique of performing an architecture search for machine learning models while respecting constraints provided by users, when combined with knowledge about performing the user's desired task.”
Claim 14 is rejected under 35 U.S.C. § 103 as being unpatentable over Deng in view of Appel in view of Wang et al. (US 2021/0241177 A1) (hereinafter “Wang”).
Regarding claim 14:
Deng in view of Appel teaches [a]n information processing system comprising: …[t]he information processing apparatus of claim 1
a terminal device; and… [t]he information processing apparatus being configured to communicate with the terminal device (Deng, page 1-2, col 2, ¶2 “Therefore, client selection, i.e., selecting appropriate mobile devices from candidate clients to participate in distributed learning, becomes crucial for high-quality federated learning.”),
Deng in view of Appel does not teach “the terminal device including terminal device processing circuitry configured to: receive a setting operation for setting conditions related to the model;
transmit condition information indicating the conditions set by the setting operation to [t]he information processing apparatus; and
cause a display to display the model specifying information received from [t]he information processing apparatus”
However, Wang teaches the terminal device including terminal device processing circuitry configured to: receive a setting operation for setting conditions related to the model (Wang, ¶129 “In one embodiment, a fourth operation entrance independent from the first operation entrance and the second operation entrance is further provided, and the fourth operation entrance is used to perform configuration regarding the providing of the prediction service by using machine learning model.” Here the sent configuration regarding the model can be considered the setting conditions related to the model);
transmit condition information indicating the conditions set by the setting operation to [t]he information processing apparatus (Wang, ¶132, “Step S9400, configuration information input through the fourth operation entrance is obtained.” Here the obtaining of the configuration information can be considered the transmitting of condition information indicating the conditions set by the setting operation); and
cause a display to display the model specifying information received from [t]he information processing apparatus (Wang, ¶150 “In the graphical interface shown in FIG. 3, the first operation entrance may be the “enter” button 401 corresponding to the behavioral data at the upper left of the GUI, and the information display area corresponding to the first operation entrance may be information displayed above the “enter” button 401; the second operation entrance may be the “enter” button 402 corresponding to the feedback data at the upper right of the GUI, and the information display area corresponding to the second operation entrance may be information displayed above the “enter” button 402; the third operation entrance may be the “enter” button corresponding to model training at the bottom right of the GUI, and the information display area corresponding to the third operation entrance may be the information displayed above the “enter” button corresponding to model training, and, the fourth operation entrance may be the “enter” button corresponding to model application at the bottom left of the GUI, and the information display area corresponding to the fourth operation entrance may be information displayed above the “entry” button corresponding to model application.” Here, the displayed information corresponding to model application can be considered the model specifying information).
Deng in view of Appel and Wang are analogous art because both references concern methods for machine learning model personalization and distribution. Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Deng/Appel’s federated learning system to incorporate the GUI taught by Wang. The motivation for doing so would have been to incorporate user-interaction and feedback as stated in Wang, ¶128 “According to the method of the embodiment, it provides an operation entrance for collecting the behavioral data and an operation entrance for collecting the feedback data, respectively, so as to import the behavioral data and feedback data into the system respectively, so that users may complete the auto-training processes of machine learning models in an easy-to-understand interactive manner.”
Claim 15 is rejected under 35 U.S.C. § 103 as being unpatentable over Deng in view of Wang.
Regarding claim 15:
Deng teaches [a]n information processing system comprising processing circuitry configured to:
receive a setting operation for setting conditions related to a model (Deng, page 6, col 1, ¶1 “To facilitate high-quality client selection within a limited budget, following the principles in Section 3, we set the features xi of each client Ci as a 4-dimensional vector represented by
x
i
=
ⅆ
i
,
q
i
l
,
q
i
d
,
b
i
, where di are the number of data samples used for training,
q
i
l
and
q
i
d
are the data quality in terms of data labels and data distribution, respectively, and bi is the required price (i.e., payment) for client Ci to complete the learning task.” Here, the state received from each client relating to its local model can be considered the setting condition indicating conditions related to a model);
determine whether at least one of a plurality of models satisfies the conditions set by the setting operation, (Deng, page 3, col 1, section 2.1, ¶1, “Therefore, to achieve satisfying FL performance, the platform needs to select an optimal subset of clients from C within the budget B for the given FL task. The selected clients can collaboratively train the FL model using their local data samples and then receive their declared payments.” Here, models on clients within the selected optimal subset can be considered the models that satisfy the conditions ), the plurality of models being each generated by executing learning processing using learning data (Deng, page 3, col 2, ¶3 “In each round r, based on the global model
w
r
, each participating client conducts model training individually by using the local data set Di, the training results of which can be used to update the local model parameters
w
i
r
.”);
Deng does not teach “cause a display to display model specifying information that specifies one of a plurality of models satisfying the conditions set by the setting operation”
However, Wang teaches cause a display to display model specifying information that specifies one of a plurality of models satisfying the conditions set by the setting operation (Wang, ¶150 “In the graphical interface shown in FIG. 3, the first operation entrance may be the “enter” button 401 corresponding to the behavioral data at the upper left of the GUI, and the information display area corresponding to the first operation entrance may be information displayed above the “enter” button 401; the second operation entrance may be the “enter” button 402 corresponding to the feedback data at the upper right of the GUI, and the information display area corresponding to the second operation entrance may be information displayed above the “enter” button 402; the third operation entrance may be the “enter” button corresponding to model training at the bottom right of the GUI, and the information display area corresponding to the third operation entrance may be the information displayed above the “enter” button corresponding to model training, and, the fourth operation entrance may be the “enter” button corresponding to model application at the bottom left of the GUI, and the information display area corresponding to the fourth operation entrance may be information displayed above the “entry” button corresponding to model application.” Here, the displayed information corresponding to model application can be considered the model specifying information).
Deng and Wang are analogous art because both references concern methods for machine learning model personalization and distribution. Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Deng’s federated learning system to incorporate the GUI taught by Wang. The motivation for doing so would have been to incorporate user-interaction and feedback as stated in Wang, ¶128 “According to the method of the embodiment, it provides an operation entrance for collecting the behavioral data and an operation entrance for collecting the feedback data, respectively, so as to import the behavioral data and feedback data into the system respectively, so that users may complete the auto-training processes of machine learning models in an easy-to-understand interactive manner.”
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Sun et al. (“FedMSA: AModelSelection and Adaptation System for Federated Learnin”, Sun et al., 24 September 2022) discloses a model selection and fast adaptation system which reduces the complexity of FL system deployment by providing automation of adaptation and deployment of training tasks as microservices in a FL life-cycle along with an optimal model selection algorithm.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JACOB Z SUSSMAN MOSS whose telephone number is (571) 272-1579. The examiner can normally be reached Monday - Friday, 9 a.m. - 5 p.m. ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kakali Chaki can be reached on (571) 272-3719. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/J.S.M./Examiner, Art Unit 2122
/KAKALI CHAKI/Supervisory Patent Examiner, Art Unit 2122
Prosecution Insights