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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 03/10/2023 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Election/Restrictions
Applicant’s election without traverse of 1-2, 6-8, and 10-22 in the reply filed on 04/28/2026 is acknowledged.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 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 1-2, 6-8, 10-14, 17, and 19-22 are rejected under 35 U.S.C. 103 as being unpatentable over Streit et al. (US-20220147607-A1) in view of Semichev et al. (US-20220084371-A1).
Regarding Claim 1,
Streit (US 20220147607 A1) teaches a learning apparatus for use in a hiding process using a neural network, comprising:
at least one processor configured to perform operations of:
obtaining output data from a first inference model that has been trained and performs predetermined processing on input data (para [0074] “Helper networks are trained on good and bad data samples to validate good samples and filter bad samples.” Helper network 204 (i.e., first inference model) is trained on input data. para [0075] “In various embodiments, bad identification data can be filtered (206) before the identification information is used for generating encrypted feature vectors at 210 and classification networks are trained. In other embodiments, good identification data can be validated to produce filtered plaintext identification information at 206.” Filtered plaintext 206 (i.e., output data).);
obtaining a second inference model for hiding the output data (para [0073] “In various settings, helper networks (204) protect embedding networks (208) by ensuring that only good identification data is used (206) to construct encrypted feature vectors (210) used to train classification networks (212) that allow identification (214).” Embedding network (i.e., second inference model) encrypts the output data.); and
using the output data from the first inference model as input data to train the second inference model (para [0028] “The method comprises instantiating, by at least one processor, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information.” Plaintext (i.e. output first model) used to train embedding network (i.e., second inference model).).
Streit does not explicitly disclose
obtaining a second inference model including a processing layer for hiding comprising at least one layer for hiding the output data;
However, Semichev (US 20220084371 A1) teaches
obtaining a second inference model including a processing layer for hiding comprising at least one layer for hiding the output data (para [0027] “The bidirectional RNN model may employ an encoding layer 112 to encrypt the input vector into a global latent vector 114.”);
Streit and Semichev are analogous because they are directed towards generating encrypted vectors using machine learning models.
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit with the encryption architecture of Semichev.
Doing so would allow for measuring the level of security of the encryption model (Semichev para [0067]).
Regarding Claim 2,
Streit and Semichev teach the learning apparatus according to claim 1. Semichev further teaches wherein the at least one processor is configured to perform further operations of:
obtaining a third inference model including a processing layer for decryption comprising at least one layer for decrypting output data from the second inference model (para [0027] “During training of the bidirectional RNN, once a global latent vector 114 is calculated, a decoding layer 116 of the bidirectional RNN model may decrypt the global latent vector into an output vector (e.g. output paragraph vector 110B).”); and
using the output data from the second inference model as input data and using the output data from the first inference model as training data to train the third inference model (para [0027] “During training of the bidirectional RNN, once a global latent vector 114 is calculated, a decoding layer 116 of the bidirectional RNN model may decrypt the global latent vector into an output vector (e.g. output paragraph vector 110B).”).
Streit and Semichev are analogous because they are directed towards generating encrypted vectors using machine learning models.
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit with the encryption architecture of Semichev.
Doing so would allow for measuring the level of security of the encryption model (Semichev para [0067]).
Regarding Claim 6,
Streit and Semichev teach the learning apparatus according to claim 1. Streit further teaches wherein the at least one processor is configured to perform further operations of: when training the second inference model using the output data from the obtained first inference model as input data, training the second inference model using training data comprising values that do not include a predetermined number of zeros or more (para [0028] “The method comprises instantiating, by at least one processor, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information. No predetermined zeros.).
Regarding Claim 7,
Streit and Semichev teach the learning apparatus according to claim 1. Streit further teaches wherein the second inference model comprises a plurality of models corresponding to the first inference model (para [0073] “According to various embodiments, plaintext identification information 202 (e.g., behavioral, biometric, physiologic, and/or digital activity based information, among other options) can be processed by helper networks (e.g., 204) that are configured to protect embedding networks (208) that generate encrypted features vectors (210).” Plurality of embedding networks.).
Regarding Claim 8,
Streit and Semichev teach the learning apparatus according to claim 1. Semichev further teahces wherein the third inference model includes the first inference model (para [0027] “The first global latent vector 114 may be decrypted by a decoder of the bidirectional recurrent neural network 116 to form an output paragraph vector 110B.” Both the encoder and decoder are bidirectional RNNs.).
Streit and Semichev are analogous because they are directed towards generating encrypted vectors using machine learning models.
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit with the encryption architecture of Semichev.
Doing so would allow for measuring the level of security of the encryption model (Semichev para [0067]).
Regarding Claim 10,
Streit (US 20220147607 A1) teaches an inference apparatus for use in a hiding process using a neural network, comprising: at least one processor configured to perform operations of:
executing a first inference model that has been trained and performs predetermined processing on input data, and obtaining output data from the first inference model (para [0074] “Helper networks are trained on good and bad data samples to validate good samples and filter bad samples.” Helper network 204 (i.e., first inference model) is trained on input data. para [0075] “In various embodiments, bad identification data can be filtered (206) before the identification information is used for generating encrypted feature vectors at 210 and classification networks are trained. In other embodiments, good identification data can be validated to produce filtered plaintext identification information at 206.” Filtered plaintext 206 (i.e., output data).);
obtaining a second inference model that has been trained for hiding the output data (para [0073] “In various settings, helper networks (204) protect embedding networks (208) by ensuring that only good identification data is used (206) to construct encrypted feature vectors (210) used to train classification networks (212) that allow identification (214).” Embedding network (i.e., second inference model) encrypts the output data.); and
using the output data from the first inference model as input data to run the second inference model (para [0028] “The method comprises instantiating, by at least one processor, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information.” Plaintext (i.e. output first model) used to train embedding network (i.e., second inference model).) and outputting output data from the second inference model as an inference result (para [0022] “According to one embodiment, the method further comprises identifying the user based on geometric evaluation of encrypted feature vectors and prediction by at least one classification network.”).
Streit does not explicitly disclose
obtaining a second inference model that has been trained and includes a processing layer for hiding comprising at least one layer for hiding the output data;
However, Semichev (US 20220084371 A1) teaches
obtaining a second inference model that has been trained and includes a processing layer for hiding comprising at least one layer for hiding the output data (para [0027] “The bidirectional RNN model may employ an encoding layer 112 to encrypt the input vector into a global latent vector 114.”);
Streit and Semichev are analogous because they are directed towards generating encrypted vectors using machine learning models.
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit with the encryption architecture of Semichev.
Doing so would allow for measuring the level of security of the encryption model (Semichev para [0067]).
Regarding Claim 11,
Streit and Semichev teach the inference apparatus according to claim 10. Streit further teaches wherein the at least one processor is configured to perform further operations of:
determining whether or not hiding by the second inference model is necessary (para [0095] “If the identification information is validated as a good sample, 404YES, process 400 continues with generation of encrypted feature vectors/embeddings at 408. The embeddings are then classified at 410 to determine any match to an identity. If there is no match locally (i.e., the classification neural network is not trained to identify the embeddings), at 412NO, process 400 can continue with a remote identification attempt at 414 using the embeddings.”); and
in a case where it is determined that the hiding is unnecessary, outputting the output data from the first inference model as an inference result without running the second inference model (para [0094] “Helper networks are configured to process identification information as an input and validate a good information sample. Here, a good sample is identified based on trained network characteristics where the helper network is trained to identify information samples that will improve identification entropy of subsequent neural networks.”).
Regarding Claim 12,
Streit and Semichev teach the inference apparatus according to claim 11. Streit further teaches wherein the at least one processor is configured to perform further operations of: determining that hiding by the second inference model is unnecessary when at least one of the following cases apply:
a case where the output data from the first inference model is recorded on a recording medium which the inference apparatus has, a case where the output data from the first inference model includes no personal information (para [0018] “According to one embodiment, the at least one processor is further configured to return an identity responsive to geometric matching executed on encrypted feature vectors generated from an input of plaintext identifying information for the entity against stored encrypted feature vectors.”), and a case where an output destination of the inference result from the inference apparatus is connected by wire (para [0006] “The system comprises at least one processor operatively connected to a memory, the at least one processor configured to: instantiate, at a local device, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information, instantiate, at the local device,”).
Regarding Claim 13,
Streit and Semichev teach the inference apparatus according to claim 11. Streit further teaches wherein the at least one processor is configured to perform further operations of: determining that hiding by the second inference model is necessary in a case where output data from the inference apparatus is to be output to an external device using a network line (para [0006]).
Regarding Claim 14,
Streit and Semichev teach the inference apparatus according to claim 11. Semichev further teaches wherein the at least one processor is configured to perform further operations of: adding identification information, which indicates whether the inference result from the inference apparatus is a first inference result comprising the output data from the second inference model (para [0027] “During training of the bidirectional RNN, once a global latent vector 114 is calculated, a decoding layer 116 of the bidirectional RNN model may decrypt the global latent vector into an output vector (e.g. output paragraph vector 110B).”) or a second inference result comprising the output data from the first inference model, to the inference result from the inference apparatus (para [0027] “The bidirectional RNN model may employ an encoding layer 112 to encrypt the input vector into a global latent vector 114.”).
Streit and Semichev are analogous because they are directed towards generating encrypted vectors using machine learning models.
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit with the encryption architecture of Semichev.
Doing so would allow for measuring the level of security of the encryption model (Semichev para [0067]).
Regarding Claim 17,
Streit (US 20220147607 A1) teaches an inference system for use in a hiding process using a neural network, comprising a learning apparatus, a first inference apparatus, and a second inference apparatus, wherein the learning apparatus comprises at least one processor configured to perform operations of:
obtaining output data from a first inference model that has been trained and performs predetermined processing on input data (para [0074] “Helper networks are trained on good and bad data samples to validate good samples and filter bad samples.” Helper network 204 (i.e., first inference model) is trained on input data. para [0075] “In various embodiments, bad identification data can be filtered (206) before the identification information is used for generating encrypted feature vectors at 210 and classification networks are trained. In other embodiments, good identification data can be validated to produce filtered plaintext identification information at 206.” Filtered plaintext 206 (i.e., output data).);
obtaining a second inference model for hiding the output data (para [0073] “In various settings, helper networks (204) protect embedding networks (208) by ensuring that only good identification data is used (206) to construct encrypted feature vectors (210) used to train classification networks (212) that allow identification (214).” Embedding network (i.e., second inference model) encrypts the output data.);
using the output data from the first inference model as input data to train the second inference model (para [0028] “The method comprises instantiating, by at least one processor, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information.” Plaintext (i.e. output first model) used to train embedding network (i.e., second inference model).);
using the output data from the second inference model as input data and using the output data from the first inference model as training data to train the third inference model (para [0073] “According to various embodiments, plaintext identification information 202 (e.g., behavioral, biometric, physiologic, and/or digital activity based information, among other options) can be processed by helper networks (e.g., 204) that are configured to protect embedding networks (208) that generate encrypted features vectors (210). In various settings, helper networks (204) protect embedding networks (208) by ensuring that only good identification data is used (206) to construct encrypted feature vectors (210) used to train classification networks (212) that allow identification (214).” Classification network (i.e., third inference model).), wherein the first inference apparatus comprises at least one processor configured to perform operations of:
running a first inference model that has been trained and performs predetermined processing on input data, and obtaining output data from the first inference model (para [0074] “Helper networks are trained on good and bad data samples to validate good samples and filter bad samples.”);
obtaining the trained second inference model from the learning apparatus (para [0006] “The system comprises at least one processor operatively connected to a memory, the at least one processor configured to: instantiate, at a local device, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information,”);
using the obtained output data from the first inference model as input data to run the obtained second inference model and outputting output data from the second inference model (para [0028] “The method comprises instantiating, by at least one processor, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information.” Plaintext (i.e. output first model) used to train embedding network (i.e., second inference model).); and
sending the output data from the second inference model to the second inference apparatus (para [0010] “According to one aspect a method for private identity is provided. The method comprises: instantiating, by at least one processor at a local device, at least one pre-trained embedding network configured to generate encrypted feature vectors from an input of plaintext identifying information, instantiating, by the least one processor at the local device, at least one local classification network, accepting, by the at least one local classification network, the encrypted feature vectors and return a matching label to an identity or an unknown result during prediction, instantiating, by at least one processor at a remote device,”), wherein the second inference apparatus comprises at least one processor configured to perform operations of:
receiving the output data from the second inference model from the first inference apparatus (para [0010]);
obtaining the trained third inference model from the learning apparatus (para [0010] “managing the at least one local classification network and remote classification network to output matching labels responsive to input of matching encrypted feature vectors.” Classification network (i.e., third inference model).); and
Streit does not explicitly disclose
obtaining a second inference model including a processing layer for hiding comprising at least one layer for hiding the output data;
obtaining a third inference model including a processing layer for decryption comprising at least one layer for decrypting output data from the second inference model; and
using the output data received from the second inference model as input data to run the obtained third inference model and decrypting the output data from the second inference model.
However, Semichev (US 20220084371 A1) teaches
obtaining a second inference model including a processing layer for hiding comprising at least one layer for hiding the output data (para [0027] “The bidirectional RNN model may employ an encoding layer 112 to encrypt the input vector into a global latent vector 114.”);
obtaining a third inference model including a processing layer for decryption comprising at least one layer for decrypting output data from the second inference model (para [0027] “During training of the bidirectional RNN, once a global latent vector 114 is calculated, a decoding layer 116 of the bidirectional RNN model may decrypt the global latent vector into an output vector (e.g. output paragraph vector 110B).”); and
using the output data received from the second inference model as input data to run the obtained third inference model and decrypting the output data from the second inference model (para [0027] “During training of the bidirectional RNN, once a global latent vector 114 is calculated, a decoding layer 116 of the bidirectional RNN model may decrypt the global latent vector into an output vector (e.g. output paragraph vector 110B).”).
Streit and Semichev are analogous because they are directed towards generating encrypted vectors using machine learning models.
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit with the encryption architecture of Semichev.
Doing so would allow for measuring the level of security of the encryption model (Semichev para [0067]).
Regarding Claim 19,
Claim 19 is the method corresponding to the apparatus of claim 1. Claim 19 is substantially similar to claim 1 and is rejected on the same grounds.
Regarding Claim 20,
Claim 20 is the method corresponding to the apparatus of claim 10. Claim 20 is substantially similar to claim 10 and is rejected on the same grounds.
Regarding Claim 21,
Claim 21 is the storage medium corresponding to the apparatus of claim 1. Claim 21 is substantially similar to claim 1 and is rejected on the same grounds.
Regarding Claim 22,
Claim 22 is the storage medium corresponding to the apparatus of claim 10. Claim 22 is substantially similar to claim 10 and is rejected on the same grounds.
Claims 15-16 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Streit/Semichev, as applied above, and further in view of Patrich et al. (US-20200151326-A1).
Regarding Claim 15,
Streit and Semichev teach the inference apparatus according to claim 14.
Streit and Semichev do not explicitly disclose
wherein the at least one processor is configured to perform further operations of: when obtaining the second inference model, selecting the second inference model from a plurality of inference models according to a security level of communication means for use in outputting the inference result.
However, Patrich (US 20200151326 A1) teaches
wherein the at least one processor is configured to perform further operations of: when obtaining the second inference model, selecting the second inference model from a plurality of inference models according to a security level of communication means for use in outputting the inference result (para [0068] “In some embodiments, the process of generating the recommendation may be performed by comparing the new security alert's attributes to attributes of other previous security alerts to identify a set of previous security alerts that are most closely related to the new security alert (e.g., a threshold number of attributes are sufficiently common or related) and then selecting a compatible prediction model based on this comparison analysis.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit and Semichev with the security threshold of Patrich.
Doing so would allow for selecting a compatible prediction model for performing analysis (Patrich para [0068]).
Regarding Claim 16,
Streit, Semichev, and Patrich teach the inference apparatus according to claim 15. Semichev further teaches wherein the identification information includes information identifying which one of the inference models is the second inference model in a case where the inference result from the inference apparatus is the first inference result (para [0027] “The bidirectional RNN model may employ an encoding layer 112 to encrypt the input vector into a global latent vector 114.”).
Streit and Semichev are analogous because they are directed towards generating encrypted vectors using machine learning models.
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit with the encryption architecture of Semichev.
Doing so would allow for measuring the level of security of the encryption model (Semichev para [0067]).
Regarding Claim 18,
Streit and Semichev teach the inference system according to claim 17.
Streit and Semichev do not explicitly disclose
wherein the at least one processor of the first inference apparatus is configured to perform further operations of: when obtaining the trained second inference model from the learning apparatus, selecting the second inference model from a plurality of inference models according to a security level.
However, Patrich (US 20200151326 A1) teaches
wherein the at least one processor of the first inference apparatus is configured to perform further operations of: when obtaining the trained second inference model from the learning apparatus, selecting the second inference model from a plurality of inference models according to a security level (para [0068] “In some embodiments, the process of generating the recommendation may be performed by comparing the new security alert's attributes to attributes of other previous security alerts to identify a set of previous security alerts that are most closely related to the new security alert (e.g., a threshold number of attributes are sufficiently common or related) and then selecting a compatible prediction model based on this comparison analysis.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the ML model of Streit and Semichev with the security threshold of Patrich.
Doing so would allow for selecting a compatible prediction model for performing analysis (Patrich para [0068]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HENRY K NGUYEN whose telephone number is (571)272-0217. The examiner can normally be reached Mon - Fri 7:00am-4:30pm.
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, Li B Zhen can be reached at 5712723768. 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.
/HENRY NGUYEN/Examiner, Art Unit 2121
Prosecution Insights