DETAILED ACTION
Response to Amendment
The amendment filed on 1 June 2026 has been entered.
Claims 1-20 are pending.
Claim 12 is cancelled.
Claims 1-11, 17-20 are amended.
Claims 1-11, 13-20 will be pending.
Response to Arguments
Applicant’s remarks, regarding the rejections of claims under 35 USC 103, have been fully considered.
Applicant respectfully submits that Gharibi combined with Vaikuntanathan and Xiong does not disclose, teach, or suggest, "partitioning, into a client-side portion and a server-side portion, a trained neural network based on a specified ratio of layers between a first set of layers and a second set of layers of the trained neural network, the client-side portion comprising the first set of layers of the trained neural network, the server-side portion comprising the second set of layers of the trained neural network, the trained neural network trained using a first set of training data" and "sending, by the server-side portion, the homomorphically encrypted intermediate result to the client-side portion, wherein the server-side portion decrypts the homomorphically encrypted intermediate result", as required by Applicant's amended Claim 1 (emphasis supplied).
Applicant’s arguments have been considered, but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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 .
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-6, 8-10, 13-17, 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Gharibi et al. (U.S. Pre-Grant Publication No. 20230306254, hereinafter ‘Gharibi'), in view of Ha et al. (NPL: "Spatio-Temporal Split Learning for Privacy-Preserving Medical Platforms: Case Studies With COVID-19 CT, X-Ray, and Cholesterol Data", hereinafter 'Ha') and Xiong et al. (U.S. Patent No. 12001577, hereinafter 'Xiong').
Regarding claim 1 and analogous claims 10, 20, Gharibi teaches A method comprising: partitioning, into a client-side portion and a server-side portion, a trained neural network, based on a specified ratio of layers between a first set of layers and a second set of layers of the trained neural network, the client-side portion comprising the first set of layers of the trained neural network, the server-side portion comprising the second set of layers of the trained neural network ([0041] FIG. 2 illustrates a split learning centralized approach. A a trained neural network model (neural network) 204 is partitioning, into a client-side portion and a server-side portion split into two parts: the client-side portion comprising the first set of layers of the trained neural network one part (206A, 208A, 210A) resides on the respective client side 206, 208, 210 and includes the input layer to the model and optionally other layers up to a cut layer, and the the server-side portion comprising the second set of layers of the trained neural network other part (B) resides on the server side 202 and often includes the output layer. Split layer (S) refers to the layer (the cut layer) where A and B are split. In FIG. 2 , SA represents a split layer or data sent from A to B and SB represents a split layer sent from B to A.),
the trained neural network trained using a first set of training data ([0079] Next is discussed a new approach to training which uses different types of training data together to train a neural network, using the blind learning approaches disclosed herein.);
receiving, from the client-side portion, a homomorphically encrypted intermediate result input ([0063] Each client first encrypts their model using homomorphic encryption and then sends the encrypted Ai′ data to the server 402.; [0074] In another aspect, a customer could choose SA, SB lines (vectors and numbers) which represent weights that need to be propagated. If a client wanted their data to be locked down without the server knowing anything about the data, that data can be homomorphically encrypted. The encryption process (which can include any encryption process) could be used in any approach disclosed above.);
Gharibi fails to teach based on a specified ratio of layers between a first set of layers and a second set of layers of the trained neural network, computing, from the homomorphically encrypted intermediate result input to the server-side portion, by the server side portion, a homomorphically encrypted output of the trained neural network, a homomorphically encrypted intermediate result comprising the homomorphically encrypted output computed by the client-side portion; and sending, by the server-side portion, the homomorphically encrypted intermediate result to the client-side portion, wherein the server-side portion decrypts the homomorphically encrypted intermediate result.
Ha teaches A method comprising: partitioning, into a client-side portion and a server-side portion, a trained neural network ([A. SYSTEM MODEL, pg. 121048] The overall architecture of our proposed system is depicted in Fig. 1. In the figure, the learning layers are partitioning split in two parts: the into a client-side portion hospitals (i.e., clients) and and a server-side portion centralized server. In our environment, hospitals are the clients that make requests to build a a trained neural network deep neural network model with its patient data. The hospital holds the input layer and the first hidden layer, and the centralized server has the rest of the layers; thus most of the computation occurs at the central server, where it builds up a high-accuracy medical neural network.),
based on a specified ratio of layers between a first set of layers and a second set of layers of the trained neural network ([2) MODEL CONFIGURATION FOR IMAGE DATA, pg. 121053] After the pre-processing step of scaling, the neural network is established. The COVID-19 chest CT scans are trained using a custom model with based on a specified ratio of layers between a first set of layers and a second set of layers of the trained neural network a total of 5 convolutional layers. In using our spatio-temporal split learning algorithm, the client will train up to the first layer, and the server will continue training along 4 of the convolutional layers.; Ha teaches partitioning the neural network between the client and server by specifying the client will train a ratio of 1/5 layers and the server will train a ratio of 4/5 layers.),
Gharibi and Ha are considered to be analogous to the claimed invention because they are in the same field of machine learning. In view of the teachings of Gharibi, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Ha to Gharibi before the effective filing date of the claimed invention in order to detach the privacy preserving layer from the rest of the learning process to minimize privacy breaches (cf. Ha, [ABSTRACT] Machine learning requires a large volume of sample data, especially when it is used in high accuracy medical applications. However, patient records are one of the most sensitive private information that is not usually shared among institutes. This paper presents spatio-temporal split learning, a distributed deep neural network framework, which is a turning point in allowing collaboration among privacy-sensitive organizations. Our spatio-temporal split learning presents how distributed machine learning can be efficiently conducted with minimal privacy concerns. The proposed split learning consists of a number of clients and a centralized server. Each client has only has one hidden layer, which acts as the privacy-preserving layer, and the centralized server comprises the other hidden layers and the output layer. Since the centralized server does not need to access the training data and trains the deep neural network with parameters received from the privacy-preserving layer, privacy of original data is guaranteed. We have coined the term, spatio-temporal split learning, as multiple clients are spatially distributed to cover diverse datasets from different participants, and we can temporally split the learning process, detaching the privacy preserving layer from the rest of the learning process to minimize privacy breaches.).
Xiong teaches computing, from the homomorphically encrypted intermediate result input to the server-side portion, by the server side portion, a homomorphically encrypted output of the trained neural network ([Col. 13, Lines 12-21] The client server runs its own local data on its client neural network and from the homomorphically encrypted intermediate result input sends its partially encrypted neural network weight updates (612, 616, 620) to the central server 602. The computing to the server-side portion, by the server side portion, a homomorphically encrypted output of the trained neural network central server 602 then processes and aggregates all partially encrypted neural network updates sent from the client servers and updates the neural network weights of the global model. The central server 602 sends the updated neural network weights back to the client servers so the client servers can continue local neural network use and training.),
a homomorphically encrypted intermediate result comprising the homomorphically encrypted output computed by the client-side portion ([Col. 14, Lines 24-43] In an embodiment, the process 700 comprises obtaining 702 input data to be processed using a machine learning model, the obtained input data being in plaintext form, such as described above. The input data can be, for example, inputs 208 described above in connection with FIG. 2. With the obtained input data, the process 700 can comprise using 704 a first portion of the machine learning model to calculate plaintext intermediate data, where the first portion of the machine learning model is in plaintext form and a second portion of the machine learning model is in ciphertext form according to a homomorphic encryption scheme, such as described above. The plaintext intermediate data can be, for example, outputs 218 described above in computed by the client-side portion connection with FIG. 2. a homomorphically encrypted intermediate result With the plaintext intermediate data obtained, the process 700 can comprise using 706 the homomorphic encryption scheme to encrypt the plaintext intermediate data to obtain ciphertext intermediate data. The ciphertext intermediate data can be, for example, comprising the homomorphically encrypted output outputs of the homomorphic cipher function 220 described above in connection with FIG. 2.); and
sending, by the server-side portion, the homomorphically encrypted intermediate result to the client-side portion, wherein the server-side portion decrypts the homomorphically encrypted intermediate result ([Col. 13, Lines 15-26] The central server 602 then processes and aggregates all partially encrypted neural network updates sent from the client servers and updates the neural network weights of the global model. The sending, by the server-side portion, the homomorphically encrypted intermediate result to the client-side portion central server 602 sends the updated neural network weights back to the client servers so the client servers can continue local neural network use and training. The federated learning system shown in FIG. 6 can utilize various techniques described herein. For example, in some embodiments, the wherein the server-side portion decrypts the homomorphically encrypted intermediate result central server 602 has sufficient cryptographic material (e.g., a private key of a public/private key pair) to decrypt the partially encrypted local neural network weights.).
Gharibi, Ha, and Xiong are considered to be analogous to the claimed invention because they are in the same field of machine learning. In view of the teachings of Gharibi and Ha, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Xiong to Gharibi before the effective filing date of the claimed invention in order to protect proprietary machine learning models against intellectual property theft, computations can be performed by a device without providing the device the ability to determine what the underlying portions of the model are in plaintext form (cf. Xiong, [Col. 2, Lines 6-23] To protect proprietary machine learning models against intellectual property theft, various techniques described below use a homomorphic encryption scheme to encrypt a portion of a machine learning model, such as one or more layers of a neural network. Plaintext input data to be input to the encrypted portion of the model is likewise encrypted using the homomorphic encryption scheme. With the plaintext input data and portion of the machine learning model encrypted, computations can be performed over the encrypted input data and the encrypted portion of the model without decrypting either. Similarly, if output from such operations are to be input into another encrypted portion of the machine learning model, additional computations involved with the other portion of the machine learning model can be performed without decryption. In this manner, computations can be performed by a device without providing the device the ability to determine what the underlying portions of the model are in plaintext form.).
Regarding claim 2 and analogous claim 13, Gharibi, as modified by Ha and Xiong, teaches The method of claim 1 and The computer program product of claim 10, respectively.
Gharibi teaches wherein the partitioning is performed using a received partition location ([0041] Split layer (S) refers to the layer partitioning is performed using a received partition location (the cut layer) where A and B are split. In FIG. 2 , SA represents a split layer or data sent from A to B and SB represents a split layer sent from B to A.).
Gharibi, Ha, and Xiong are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 3 and analogous claim 14, Gharibi, as modified by Ha and Xiong, teaches The method of claim 1 and The computer program product of claim 10, respectively.
Xiong teaches wherein the partitioning is performed using a partition location computed using data of a system computing the homomorphically encrypted output of the trained neural network ([Col. 2, Lines 60-64] In some examples, the partitioning is performed first portion of the machine learning model comprises a set of layers of a neural network and the second portion of the machine learning model comprises a set of second layers of the neural network.; [Col. 5, Lines 31-37] As described in FIG. 1, the neural network 108 may have multiple layers and each layer comprises sequences of operation, wherein the using a partition location output of one layer and operation can be used as input to another layer and operation. The neural network 108 partially encrypts the input plaintext 106 into ciphertext output 110, which comprises both partial ciphertext and partial plaintext output data.; [Col. 8, Lines 11-18] The neural network 210, in this example, processes the inputs 208 through the first layers of the neural network to result in intermediate inputs 218, which can be output of the plaintext layers 212, sends the intermediate inputs 218 through the homomorphic cipher function 216, which results in partially encrypted input data 220. This partially computed using data of a system computing the homomorphically encrypted output of the trained neural network encrypted output data 220 is made up of plaintext output data and ciphertext output data.; [Col. 18, Lines 5-11] The data store 910, in an embodiment, is operable, through logic associated therewith, to receive instructions from the application server 908 and obtain, update or otherwise process data in response thereto, and the application server 908 provides static, dynamic, or a combination of static and dynamic data in response to the received instructions.).
Gharibi, Ha, and Xiong are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 4 and analogous claim 15, Gharibi, as modified by Ha and Xiong, teaches The method of claim 3 and The computer program product of claim 14, respectively.
Xiong teaches wherein the partition location is computed using static data of the system computing the homomorphically encrypted output of the trained neural network ([Col. 2, Lines 60-64] In some examples, the first portion of the machine learning model comprises a set of layers of a neural network and the second portion of the machine learning model comprises a set of second layers of the neural network.; [Col. 5, Lines 31-37] As described in FIG. 1, the neural network 108 may have multiple layers and each layer comprises sequences of operation, wherein the partition location output of one layer and operation can be used as input to another layer and operation. The neural network 108 partially encrypts the input plaintext 106 into ciphertext output 110, which comprises both partial ciphertext and partial plaintext output data.; [Col. 8, Lines 11-18] The neural network 210, in this example, processes the inputs 208 through the first layers of the neural network to result in intermediate inputs 218, which can be output of the plaintext layers 212, sends the intermediate inputs 218 through the homomorphic cipher function 216, which results in partially encrypted input data 220. This partially of the system computing the homomorphically encrypted output of the trained neural network encrypted output data 220 is made up of plaintext output data and ciphertext output data.; [Col. 18, Lines 5-11] The data store 910, in an embodiment, is operable, through logic associated therewith, to receive instructions from the application server 908 and obtain, update or otherwise process data in response thereto, and the application server 908 provides is computed using static data static, dynamic, or a combination of static and dynamic data in response to the received instructions.).
Gharibi, Ha, and Xiong are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 5 and analogous claim 16, Gharibi, as modified by Ha and Xiong, teaches The method of claim 3 and The computer program product of claim 14, respectively.
Xiong teaches wherein the partition location is computed using dynamic data of the system computing the homomorphically encrypted output of the trained neural network ([Col. 2, Lines 60-64] In some examples, the first portion of the machine learning model comprises a set of layers of a neural network and the second portion of the machine learning model comprises a set of second layers of the neural network.; [Col. 5, Lines 31-37] As described in FIG. 1, the neural network 108 may have multiple layers and each layer comprises sequences of operation, wherein the partition location output of one layer and operation can be used as input to another layer and operation. The neural network 108 partially encrypts the input plaintext 106 into ciphertext output 110, which comprises both partial ciphertext and partial plaintext output data.; [Col. 8, Lines 11-18] The neural network 210, in this example, processes the inputs 208 through the first layers of the neural network to result in intermediate inputs 218, which can be output of the plaintext layers 212, sends the intermediate inputs 218 through the homomorphic cipher function 216, which results in partially encrypted input data 220. This partially of the system computing the homomorphically encrypted output of the trained neural network encrypted output data 220 is made up of plaintext output data and ciphertext output data.; [Col. 18, Lines 5-11] The data store 910, in an embodiment, is operable, through logic associated therewith, to receive instructions from the application server 908 and obtain, update or otherwise process data in response thereto, and the application server 908 provides static, is computed using dynamic data dynamic, or a combination of static and dynamic data in response to the received instructions.).
Gharibi, Ha, and Xiong are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 6 and analogous claim 17, Gharibi, as modified by Ha and Xiong, teaches The method of claim 1 and The computer program product of claim 10, respectively.
Gharibi teaches further comprising: further training, using a second set of training data, the server-side portion, the further training resulting in a further trained neural network with an improved accuracy from an accuracy of the trained neural network ([0038] The training process in this case includes a server 102 creating a model 104 and sharing the model 106A, 108A and 110A with respective clients 106, 108, 110 in a linear fashion. The clients train the respective model 106A, 108A, 110A separately when they receive the model on their turn and respectively send their further training, using a second set of training data trained model data the server-side portion back to the server 102 as shown. The server 102 averages the models and produces a new model 104 with updated weights (a.k.a the further training resulting in a further trained neural network a trained model). The server 102 sends the new model or weights to the respective clients 106, 108, 110 in a linear fashion. The process is repeated a number of iterations or with an improved accuracy from an accuracy of the trained neural network until a specific accuracy is achieved.).
Gharibi, Ha, and Xiong are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 8 and analogous claim 19, Gharibi, as modified by Ha and Xiong, teaches The method of claim 6 and The computer program product of claim 17, respectively.
Gharibi teaches further comprising: second partitioning, into a second client-side portion and a second server-side portion, the further trained neural network, wherein the second client-side portion has a different number of layers from the client-side portion, wherein the second partitioning is performed responsive to determining that the improved accuracy of the further trained neural network is less than a threshold accuracy ([0090] The training of the network is done by a sequence of distributed training processes. The forward propagation and the back-propagation can take place as follows. With the raw data, a client (say client 702) trains the client-side network 702A up to a certain layer of the network, which can be called the cut layer or the split layer, and sends the activations of the cut layer to the server 710. The server 710 trains the remaining layers of the NN with the activations that it received from the client 702. This completes a single forward propagation step. second partitioning, into a second client-side portion and a second server-side portion, the further trained neural network A similar process occurs in parallel for the wherein the second client-side portion has a different number of layers from the client-side portion second client 704 and its client side network 704A and its data and generated activations which are transmitted to the server 710. A further similar process occurs in parallel for the third client 706 and its client side network 706A and its data and generated activations which are transmitted to the server 710.; [0095] As introduced above, client one 702, client two 704 and client three 706 could have different data types. The server 710 will create two parts of the network and sends one part 702A, 704A, 706A to all the clients 702, 704, 706. The system second partitioning is performed responsive to determining that the improved accuracy of the further trained neural network is less than a threshold accuracy repeats certain steps until an accuracy condition or other condition is met, such as all the clients sending data to the part of the network that they have, and sends the output to the server 710.); and
retraining, using a third set of training data, the second server-side portion, the retraining resulting in another improved accuracy improvement from the further trained neural network ([0095] As introduced above, client one 702, client two 704 and client three 706 could have different data types. The server 710 will create two parts of the network and sends one part 702A, 704A, 706A to all the clients 702, 704, 706. The system repeats certain steps until an accuracy condition or other condition is met, such as all the clients sending data to the part of the network that they have, and sends the output to the server 710. The server 710 calculates the loss value for each client and the average loss across all the clients. The retraining, using a third set of training data, the second server-side portion server 710 can update its model using a weighted average of the gradients that it computes during back-propagation and sends the gradients back to all the clients 702, 704, 706. The clients 702, 704, 706 receives the gradients from the server 710 and each client 702, 704, 706 performs the back-propagation on their client-side network 702A, 704A, 706A and computes the respective gradients for each client-side-network 702A, 704A, 706A. The respective gradients from the client-side networks 702A, 704A, 706A can then be transmitted back to the server 710 which conducts an averaging of the client-side updates and sends the global result back to all the clients 702, 704, 706.; [0050] retraining resulting in another improved accuracy improvement from the further trained neural network FIG. 4 illustrates the improvement to training neural networks disclosed herein. This improvement can be characterized as a blind learning approach and addresses some of the deficiencies of the approaches disclosed above. FIG. 4 introduces a parallel processing approach. The parallel and independent processing causes the model training to occur at a faster pace than the other models described above.).
Gharibi, Ha, and Xiong are combinable for the same rationale as set forth above with respect to claim 1.
Regarding claim 9, Gharibi, as modified by Ha and Xiong, teaches The method of claim 1.
Xiong teaches wherein the homomorphically encrypted intermediate result is computed by the client-side portion in an unencrypted form and subsequently homomorphically encrypted ([Col. 14, Lines 24-43] In an embodiment, the process 700 comprises obtaining 702 input data to be processed using a machine learning model, the obtained input data being in plaintext form, such as described above. The input data can be, for example, inputs 208 described above in connection with FIG. 2. With the obtained input data, the process 700 can comprise using 704 a first portion of the machine learning model to calculate plaintext intermediate data, where the first portion of the machine learning model is in plaintext form and a second portion of the machine learning model is in ciphertext form according to a homomorphic encryption scheme, such as described above. The plaintext intermediate data can be, for example, outputs 218 described above in connection with FIG. 2. wherein the homomorphically encrypted intermediate result is computed by the client-side portion in an unencrypted form With the plaintext intermediate data obtained, the process 700 can comprise using 706 the homomorphic encryption scheme to and subsequently homomorphically encrypted encrypt the plaintext intermediate data to obtain ciphertext intermediate data. The ciphertext intermediate data can be, for example, outputs of the homomorphic cipher function 220 described above in connection with FIG. 2.).
Gharibi, Ha, and Xiong are combinable for the same rationale as set forth above with respect to claim 1.
Claims 7, 18 are rejected under 35 U.S.C. 103 as being unpatentable over Gharibi, in view of Ha, Xiong, and further in view of Barham et al. (U.S. Pre-Grant Publication No. 20200228339, hereinafter 'Barham').
Regarding claim 7 and analogous claim 18, Gharibi, as modified by Ha and Xiong, teaches The method of claim 6 and The computer program product of claim 17, respectively.
Gharibi, as modified by Ha and Xiong, fails to teach wherein the first set of training data is public, and the second set of training data is non-public.
Barham teaches wherein the first set of training data is public, and the second set of training data is non-public ([0016] In embodiments, the neural network classifier, or a wherein the first set of training data is public subset of the layers of the neural network classifier may be trained using publicly-available non-private biometric information, and and the second set of training data is non-public other layers of the neural network classifier may be re-trained using private biometric information of a specific individual. For example, to reduce the memory and speed requirements, the whole Neural-Network model may be trained using publicly available biometric data (that doesn't have any privacy constraints). The weights of, for example, the first few layers may be fixed or store and then, during enrollment, the remaining layers may be retrained using private, person-specific, biometric data. Thus, for example, only the last few layers need be encrypted. During verification, the biometric features may be first fed to the not-encrypted NN layers, then they may be encrypted and sent to the encrypted model. The trained weights may be encrypted 210 using, for example, homomorphic encryption and transmitted 114 to server 102, which may store the encrypted neural network 108 for that person. Homomorphic encryption allows computation on encrypted data such that when the results of the computation on the encrypted data is decrypted, the results are the same as if the computation had been performed on the unencrypted or plaintext data.).
Gharibi, Ha, Xiong, and Barham are considered to be analogous to the claimed invention because they are in the same field of machine learning. In view of the teachings of Gharibi, Ha, and Xiong, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Barham to Gharibi before the effective filing date of the claimed invention in order to provide security claims that can be measured against current cryptographic solutions, such as symmetric and asymmetric methods, since embodiments may be include known and accepted cryptographic modules (cf. Barham, [0004] Embodiments of the present systems and methods may provide encrypted biometric information that can be stored and used for authentication with undegraded recognition performance. Embodiments may provide advantages over current techniques. For example, embodiments may provide security claims that can be measured against current cryptographic solutions, such as symmetric and asymmetric methods, since embodiments may be include known and accepted cryptographic modules. Further, the degradation in recognition performance rates can be described as a trade-off with memory and speed requirements, and for industry acceptable performance requirements embodiments may achieve both.).
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Gharibi, in view of Ha, Xiong, and further in view of Rakshit et al. (U.S. Pre-Grant Publication No. 20210344995, hereinafter 'Rakshit').
Regarding claim 11, Gharibi, as modified by Ha and Xiong, teaches The computer program product of claim 10.
Gharibi, as modified by Ha and Xiong, fails to teach wherein the program instructions are stored on the one or more computer readable storage media in a data processing system, and wherein the program instructions are transferred over a network from a remote data processing system.
Rakshit teaches wherein the stored program instructions are stored in a computer readable storage device in a data processing system ([0101] While it is understood that the process software (e.g., any of the instructions stored in instructions 560 of FIG. 5 and/or any software configured to perform any portion of the method described with respect to FIGS. 2-4 and/or implement any portion of the functionality discussed in FIG. 1) can be deployed by manually loading it directly in the client, server, and proxy computers via loading a storage medium such as a CD, DVD, etc., the process software can also be automatically or semi-automatically deployed into a computer system by sending the process software to a central server or a group of central servers. The process software is then downloaded into the client computers that will execute the process software.), and
wherein the stored program instructions are transferred over a network from a remote data processing system ([0062] FIG. 5 illustrates a block diagram of an example computer 500 in accordance with some embodiments of the present disclosure. In various embodiments, computer 500 can perform any or all of the method described in FIGS. 2-4 and/or implement the functionality discussed in any one of FIG. 1. In some embodiments, computer 500 receives instructions related to the aforementioned methods and functionalities by wherein the stored program instructions are transferred over a network from a remote data processing system downloading processor-executable instructions from a remote data processing system via network 550.).
Gharibi, Ha, Xiong, and Rakshit are considered to be analogous to the claimed invention because they are in the same field of machine learning. In view of the teachings of Gharibi, Ha, and Xiong, it would have been obvious for a person of ordinary skill in the art to apply the teachings of Rakshit to Gharibi before the effective filing date of the claimed invention in order to monitor, control, and report, provide transparency for both the provider and consumer of the utilized service (cf. Rakshit, [0076] Measured service: cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service.).
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 extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MAGGIE MAIDO whose telephone number is (703) 756-1953. The examiner can normally be reached M-Th: 6am - 4pm.
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, Michael Huntley can be reached on (303) 297-4307. 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.
/MM/Examiner, Art Unit 2129
/MICHAEL J HUNTLEY/Supervisory Patent Examiner, Art Unit 2129
Prosecution Insights