CTFR 17/815,876 CTFR 99217 Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia 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 04/14/2026 in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner. Response to Amendment The amendment filed on 12/24/2025 has been accepted and considered in this office action. Claims 1, 7, 9, 10, 12, 17 and 20 have been amended. No new claims have been added. No claims have been cancelled. Response to Arguments 07-37 AIA Applicant's arguments filed 12/24/2025 have been fully considered but they are not persuasive. Under the Broadest Reasonable Interpretation (BRI) standard applied during examination, limitations from the specification cannot be read into the claims. Claim 1 recites an apparatus “decomposing raw data of a system,” “fundamental [meta]data … determined at least by underlying system behavior and configurations,” and “inference metadata … determined at least by system operational conditions.” The claim does not recite “scientific systems,” “governing laws,” “geometry,” or “differential equations.” A mobile phone, personal computer, autonomous vehicle, or virtual machine as disclosed in Mukherjee (para 21) meets the claimed “system.” Consequently, Mukherjee’s “identifying information” (e.g., IP address or a VIN)) represents network configuration data that teaches the claimed “fundamental metadata” because it represents “underlying system behavior and configurations”, one an example of dynamic configuration information (an IP address) and one an example of a static quasi-permanent configuration information (a VIN). Mukherjee’s “non-identifying information” (e.g., battery behavior, vehicle environment data) teaches “inference metadata” because it which represents the operational conditions of the device and environment that change over time. Applicant argues that Mukherjee fails to disclose any processing act to “decompose” raw data. The applicant’s argument is unpersuasive. Mukherjee (para 11, 21, 32) teaches receiving data that includes both identifying information and non-identifying information, and subsequently processing this data to obfuscate the identifying portion while retaining the non-identifying portion. Mukherjee implies that the system isolates identifying fields (e.g., IP address, IMEI, device identifier, vehicle identification number (VIN), or process ID)) from the non-identifying field (e.g., battery behavior, vehicle environment data) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information teaches decomposing a set of raw data as claimed. Applicant argues that Mukherjee’s non-identifying information such as device model, OS version, email spam reporting, battery behavior, vehicle environment data, could itself fingerprint the system when aggregated. Mukherjee is not limited to obfuscating only explicitly identifying fields or only the items applicant has contentions with. (para 58-61) Mukherjee discloses obfuscating selected identifying/non-identifying information. In particular, Mukherjee discloses an example in which data includes A and D as identifying information and B and C as non-identifying information, and the obfuscator generates obfuscated data including XXA, XXB, C and XXD, where XXB is obfuscated non-identifying information. Mukherjee discloses that non-identifying information B may be obfuscated because the machine learning module does not require it. Thus, even if certain non- identifying information could fingerprint the endpoint/system, Mukherjee teaches obfuscating such selected non-identifying data when it is not required by the machine learning module. Applicant argues Mukherjee fails to obfuscate fundamental metadata responsive to “concealment operators” and “a deception kernel.” The terms “concealment operators” and “deception kernel” are broad functional terms. Under Broadest Reasonable Interpretation (BRI), a “kernel” is simply a core logical routine or process, and “operators” are mathematical or algorithmic functions. (para 34-35) Mukherjee discloses choosing between various cryptographic hashes, symmetric encryption, and tokens to obscure identifying information received from endpoints which is under BRI equivalent to generating/choosing “concealment operators.” Furthermore, (para 57) Mukherjee discloses that the obfuscator module can adapt the data generated to mee the specific requirements (manifest) of the machine learning modules. A software module (obfuscator) executing a specific logic policy (manifest) to dynamically adapt how data is obfuscated, under BRI is the deception kernel. The applicant’s argument is ultimately not persuasive . Claim Rejections - 35 USC § 112 07-30-02 AIA 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 1-20 are rejected under 35 U.S.C 112 (b) as being indefinite. The examiner finds that a person of ordinary skill in the art would not understand whether “fundamental data” refers to “fundamental metadata” in the subsequent recitation of “the fundamental data determined at least by underlying system behavior and configurations.” “the fundamental data” lacks antecedent basis and it’s unclear whether it references “the fundamental metadata” or to a different data. Because the claim later relies on this data for obfuscation and identity-concealment limitations, the scope of the claim is unclear. For the purpose of further examination, the Examiner interprets “fundamental data” being “fundamental metadata.” Claim 1, 12, and 17 are further indefinite because the phrase “underlying system behavior and configurations” contains the relative terminology of “underlying” and the disclosure fails to provide a reasonably clear boundary for the claimed subject matter. The disclosure fails to adequately distinguish, with certainty, how behavior or configuration qualify as “underlying” or how they are distinguished from “system operational conditions.” As a result, a person of ordinary skill in the art would not be able to determine whether any practicable item of metadata, such as IP address, OS version, any information from an event log and so on falls within the claimed “fundamental metadata” versus the claimed “inference metadata.” Applicant’s specification has been carefully reviewed for the concepts of “fundamental metadata” and “inference metadata”. Applicant’s specification describes “fundamental metadata” as the “governing laws” of a system, such as “physicals principles and constraints” (0024), “geometry”, “material composition”, and “differential equations”, and may include “proprietary information” that “includes the identity of the system” (0040). Yet, the disclosure then defines “inference metadata” as essentially any metadata that is not fundamental metadata (“inference metadata 112 (e.g., metadata representing anything that is not fundamental including, for example, time and external dependencies )”; [0039] ). Applicant’s description of the two metadata categories is with such vague and nebulous terminology, that the examiner finds a person of ordinary skill would not be able to consistently determine whether any particular piece of metadata is one category or the other. To put this another way, whether a piece of metadata falls within the set of “fundamental metadata”, which would be required for determining anticipation or infringement, is indeterminable. Applicant’s own arguments against the prior art rejection further evidence this issue. Claim Rejections - 35 USC § 102 07-06 AIA 15-10-15 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. 07-07-aia AIA 07-07 The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action: A person shall be entitled to a patent unless – 07-08-aia AIA (a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention. 07-12-aia AIA (a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention. 07-15 AIA Claim (s) 1-4, 6-15, 17, 20 are rejected under 35 U.S.C. 102 ( a)(1) and (a)(2 ) as being anticipated by Mukherjee (US20220092468 A1) . Regarding Claim 1, Mukherjee teaches, An apparatus, comprising: a communication terminal configured to transmit information to an artificial intelligence engine (Mukherjee, para 21, 24 discloses one or more machine learning modules 108 receives the obfuscated data 112 from the obfuscator module 104.); and a processing circuitry including one or more processors to execute computing instructions stored on a non-transitory computer readable medium, the processing circuitry configured to: decompose raw data into fundamental metadata and inference metadata (Mukherjee, para 32, discloses an example of data containing phone number, IMEI, IP address, OS version, screen on/off time. Mukherjee implies that the system then isolates identifying fields (IP address which represents network configuration data) from the non-identifying information (e.g., battery behavior, vehicle environment data) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information is functionally equivalent to decomposing a set of raw data;); the fundamental data determined at least by underlying system behavior and configurations (para 32, Mukherjee discloses identifying information (e.g., IP address)) which represents network configuration data); the inference metadata determined at least by system operational conditions (para 32, Mukherjee discloses non-identifying information (e.g., battery behavior, vehicle environment data, etc.); para 40, Mukherjee implies battery behavior represents operational condition of the battery); generate or choose one or more concealment operators (Mukherjee, para 34, 35 discloses generating token and encrypted string to conceal identifying information.); generate a deception kernel responsive to the inference metadata and the one or more concealment operators (Mukherjee, para 22, discloses the policy/configuration of obfuscator module is configured to obscure the identifying information received from the endpoint by a token (para 34) or encrypted string (para 35); It depends on both the non-identifying information that must remain original for machine learning and the specific technique (hashing/encryption) to be applied.); obfuscate the fundamental metadata responsive to the one or more concealment operators and the deception kernel to conceal identity of the system at least as indicated by the underlying system behavior and configurations (Mukherjee, para 22, discloses the obfuscator module obscures the identifying information (para 21, identification information regarding the endpoint 102 such as, for example, an IP address which represents network configuration data) received from the endpoint by a token (para 34) or encrypted string (para 35)); and provide the obfuscated fundamental metadata and the inference metadata to the artificial intelligence engine for processing (Mukherjee, para 22, 24, discloses the resulting obfuscated data 112 includes both the non-identifying information (as originally received) and obfuscated identifying information which is fed to one or more machine learning module 108 (para 24)). Regarding Claim 2, Mukherjee teaches, The apparatus of claim 1, wherein the processing circuitry is configured to generate the deception kernel responsive to the inference metadata, the one or more concealment operators, and the fundamental metadata (Mukherjee, para 22, discloses the policy/configuration of obfuscator module is configured to obscure the identifying information received from the endpoint by a token (para 34) or encrypted string (para 35); It depends on both the non-identifying information that must remain original, identifying information that needs to be obfuscated for machine learning and the specific technique (hashing/encryption) to be applied.) Regarding Claim 3, Mukherjee teaches, The apparatus of claim 2, wherein the processing circuitry is configured to obfuscate the fundamental metadata by fusing the concealment operators and the fundamental metadata together (Mukherjee, para 34, 35, discloses replacing the identifying information by a token generated using cryptographic hash or by an encrypted string. Both of these methods combine the identifying information with a cryptographic element (hash/encrypted string) to create an obfuscated data set.). Regarding Claim 4, Mukherjee teaches, The apparatus of claim 1, wherein the processing circuitry is configured to obfuscate the fundamental metadata by replacing the fundamental metadata with the concealment operators (Mukherjee, para 34, 35, discloses obfuscating the identifying information in the received data includes replacing the identifying information by a token.) . Regarding Claim 6, Mukherjee teaches, The apparatus of claim 1, wherein the communication terminal is configured to receive information from the artificial intelligence engine (Mukherjee, para 25, discloses the machine learning module 108 can transmit an actionable inference message 114 to an endpoint.). Regarding Claim 7, Mukherjee teaches, The apparatus of claim 1, wherein the processing circuitry is configured to generate or choose one or more concealment operators by decomposing a second set of raw data (Mukherjee, para 32, discloses an example of data containing phone number, IMEI, IP address, OS version, screen on/off time. Mukherjee implies that the system then isolates identifying fields (phone number, IMEI, IP address) from the non-identifying information (e.g., battery behavior, vehicle environment data) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information is functionally equivalent to decomposing a set of raw data; para 34, 35, Mukherjee discloses generating token and encrypted string to conceal identifying information.). Regarding Claim 8, Mukherjee teaches, The apparatus of claim 1, wherein the processing circuitry is configured to generate at least two sets of concealment operators, where a first set of concealment operators has a first security level and a second set of concealment operators has a second security level (Mukherjee, para 22, discloses the endpoint may choose a first privacy level or second privacy level and use either hash - token or encryption (para 34, 35) to obfuscate identifying information.). Regarding Claim 9, Mukherjee teaches, The apparatus of claim 1, wherein the processing circuitry is configured to generate or choose the one or more concealment operators responsive to first one-way hash functions (Mukherjee, para 34, discloses generating token (generated by cryptographic one-way hash) to conceal identifying information.); and generate the deception kernel responsive to second one-way hash functions (Mukherjee, para 22, discloses the policy/configuration of obfuscator module is configured to obscure the identifying information received from the endpoint by a token (generated by cryptographic one-way hash) (para 34)). Regarding Claim 10, Mukherjee teaches, The apparatus of claim 1, wherein the fundamental metadata represents underlying governing laws related to the system described by the raw data (Mukherjee, para 21, discloses the system identifies the identifying information as sensitive and needs to be concealed such as IP address, IMEI, device identifier, vehicle identification number (VIN), or process ID. These identifiers are a form of underlying, unchanging data that governs or identifies a specific system.). Regarding Claim 11, Mukherjee teaches, The apparatus of claim 1, wherein the inference metadata represents data used to train the artificial intelligence engine (Mukherjee, para 22, 24, discloses the resulting obfuscated data 112 includes both the non-identifying information (as originally received) and obfuscated identifying information which is fed to one or more machine learning module 108 (para 24) for training purposes in addition to inference derivation.) Regarding Claim 12, Mukherjee teaches, decompose raw data of a proprietary system into fundamental metadata and inference metadata (Mukherjee, para 32, discloses an example of data containing phone number, IMEI, IP address, OS version, screen on/off time. Mukherjee implies that the system then isolates identifying fields (IP address which represents network configuration data) from the non-identifying information (e.g., battery behavior, vehicle environment data) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information is functionally equivalent to decomposing a set of raw data;); the fundamental data determined at least by underlying system behavior and configurations (para 32, Mukherjee discloses identifying information (e.g., IP address)) which represents network configuration data); the inference metadata determined at least by system operational conditions (para 32, Mukherjee discloses non-identifying information (e.g., battery behavior, vehicle environment data, etc.); para 40, Mukherjee implies battery behavior represents operational condition of the battery); generate or choose one or more concealment operators (Mukherjee, para 34, 35 discloses generating token and encrypted string to conceal identifying information.); generate a deception kernel responsive to the inference metadata and the one or more concealment operators (Mukherjee, para 22, discloses the policy/configuration of obfuscator module is configured to obscure the identifying information received from the endpoint by a token (para 34) or encrypted string (para 35); It depends on both the non-identifying information that must remain original for machine learning and the specific technique (hashing/encryption) to be applied.); obfuscate the fundamental metadata responsive to the one or more concealment operators and the deception kernel to conceal identity of the system at least as indicated by the underlying system behavior and configurations (Mukherjee, para 22, discloses the obfuscator module obscures the identifying information (para 21, identification information regarding the endpoint 102 such as, for example, an IP address which represents network configuration data) received from the endpoint by a token (para 34) or encrypted string (para 35)); an artificial intelligence engine comprising computing instructions stored on a second non-transitory computer readable medium and executable by one or more second processors, the artificial intelligence engine configured to: receive data from the deception engine, the data comprising the obfuscated fundamental metadata and the inference metadata (Mukherjee, para 24 discloses one or more machine learning modules 108 receives obfuscated data 112 from the obfuscator module 104; para 22, Mukherjee discloses that the resulting obfuscated data 112 includes both the non-identifying information (as originally received) and obfuscated identifying information.); process the received data (Mukherjee, para 24 discloses that the obfuscated data 112 may be used by the one or more machine learning modules 108 for training purposes); and provide the obfuscated fundamental metadata and the inference metadata to the artificial intelligence engine for processing (Mukherjee, para 22, 24, discloses the resulting obfuscated data 112 includes both the non-identifying information (as originally received) and obfuscated identifying information which is fed to one or more machine learning module 108 (para 24)). Regarding Claim 13, Mukherjee teaches, The system of claim 12, wherein the deception engine is configured to provide the obfuscated fundamental metadata and the inference metadata to the artificial intelligence engine (Mukherjee, para 22, 24, discloses the resulting obfuscated data 112 includes both the non- identifying information (as originally received) and obfuscated identifying information which is fed to one or more machine learning module 108 (para 24) for training purposes in addition to inference derivation.) Regarding Claim 14, Mukherjee teaches, The system of claim 12, wherein the deception engine is configured to generate the deception kernel responsive to the inference metadata, the one or more concealment operators, and the fundamental metadata (Mukherjee, para 22, discloses the policy/configuration of obfuscator module is configured to obscure the identifying information received from the endpoint by a token (para 34) or encrypted string (para 35); It depends on both the non-identifying information that must remain original, identifying information that needs to be obfuscated for machine learning and the specific technique (hashing/encryption) to be applied.) Regarding Claim 15, Mukherjee teaches, The system of claim 12, wherein the artificial intelligence engine is configured to process the inference metadata of the received data (Mukherjee, para 22, 24, discloses the resulting obfuscated data 112 includes both the non-identifying information (as originally received) and obfuscated identifying information which is fed to one or more machine learning module 108 (para 24) for training purposes in addition to inference derivation.) Regarding Claim 17, Mukherjee teaches, decomposing raw data into fundamental metadata and inference metadata (Mukherjee, para 32, discloses an example of data containing phone number, IMEI, IP address, OS version, screen on/off time. Mukherjee implies that the system then isolates identifying fields (IP address which represents network configuration data) from the non-identifying information (e.g., battery behavior, vehicle environment data) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information is functionally equivalent to decomposing a set of raw data;); the fundamental data determined at least by underlying system behavior and configurations (para 32, Mukherjee discloses identifying information (e.g., IP address)) which represents network configuration data); the inference metadata determined at least by system operational conditions (para 32, Mukherjee discloses non-identifying information (e.g., battery behavior, vehicle environment data, etc.); para 40, Mukherjee implies battery behavior represents operational condition of the battery); generating or choosing one or more concealment operators (Mukherjee, para 34, 35 discloses generating token and encrypted string to conceal identifying information.); generating a deception kernel responsive to the inference metadata and the one or more concealment operators (Mukherjee, para 22, discloses the policy/configuration of obfuscator module is configured to obscure the identifying information received from the endpoint by a token (para 34) or encrypted string (para 35); It depends on both the non-identifying information that must remain original for machine learning and the specific technique (hashing/encryption) to be applied.); obfuscating the fundamental metadata responsive to the one or more concealment operators and the deception kernel to conceal identity of the system at least as indicated by the underlying system behavior and configurations (Mukherjee, para 22, discloses the obfuscator module obscures the identifying information (para 21, identification information regarding the endpoint 102 such as, for example, an IP address which represents network configuration data) received from the endpoint by a token (para 34) or encrypted string (para 35)); and providing the obfuscated fundamental metadata and the inference metadata to the artificial intelligence engine for processing (Mukherjee, para 22, 24, discloses the resulting obfuscated data 112 includes both the non-identifying information (as originally received) and obfuscated identifying information which is fed to one or more machine learning module 108 (para 24)). Regarding Claim 20, Mukherjee teaches, The method of claim 17, wherein generating or choosing one or more concealment operators comprises decomposing a second set of raw data (Mukherjee, para 32, discloses an example of data containing phone number, IMEI, IP address, OS version, screen on/off time. Mukherjee implies that the system then isolates identifying fields (phone number, IMEI, IP address) from the non-identifying information (e.g., OS version, screen on/off time) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information is functionally equivalent to decomposing a set of raw data; para 34, 35, Mukherjee discloses generating token and encrypted string to conceal identifying information.) . Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 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. 07-20-aia AIA 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. 07-21-aia AIA Claim s 5 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Mukherjee (US20220092468 A1) in view of Nagarajan (US 20150088469 A1) . Regarding Claim 5, Mukherjee teaches, The apparatus of claim 1, wherein the processing circuitry is configured to decompose the raw data into the fundamental metadata and the inference metadata (Mukherjee, para 32, discloses an example of data containing phone number, IMEI, IP address, OS version, screen on/off time. Mukherjee implies that the system then isolates identifying fields (phone number, IMEI, IP address) from the non-identifying information (e.g., OS version, screen on/off time) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information is functionally equivalent to decomposing a set of raw data); Mukherjee does not explicitly teach; However, Nagarajan teaches, passing the raw data through a reduced order model (Nagarajan, para 16 discloses feeding input data to reduced order model). It would have been obvious to a person of ordinary skills in the art before the effective filing date to modify Mukherjee’s technique of separating the identifying and non-identifying information from the raw data to incorporate the teaching of Nagarajan to implement reduced order model for the system to fed the input data. One would be motivated to make such modifications on Mukherjee’s system to effectively separate a large, complex dataset into two independent components with minimal loss of information. By doing so, Mukherjee’s system gains, an automated, proven, and computationally efficient mechanism to perform the necessary data separation process. Regarding Claim 18, Mukherjee teaches, The method of claim 17, wherein decomposing the raw data into the fundamental metadata and the inference metadata comprises (Mukherjee, para 32, discloses an example of data containing phone number, IMEI, IP address, OS version, screen on/off time. Mukherjee implies that the system then isolates identifying fields (phone number, IMEI, IP address) from the non-identifying information (e.g., OS version, screen on/off time) and processes them separately. The process of recognizing/separating the identifying information from non-identifying information is functionally equivalent to decomposing a set of raw data); Mukherjee does not explicitly teach; However, Nagarajan teaches, passing the raw data through a reduced order model (Nagarajan, para 16 discloses feeding input data to reduced order model). It would have been obvious to a person of ordinary skills in the art before the effective filing date to modify Mukherjee’s technique of separating the identifying and non-identifying information from the raw data to incorporate the teaching of Nagarajan to implement reduced order model for the system to fed the input data. One would be motivated to make such modifications on Mukherjee’s system to effectively separate a large, complex dataset into two independent components with minimal loss of information. By doing so, Mukherjee’s system gains, an automated, proven, and computationally efficient mechanism to perform the necessary data separation process . 07-21-aia AIA Claim s 16 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Mukherjee (20220092468 A1) in view of Schnabel (Schnabel, L., Matzka, S., Stellmacher, M., Patzold, M., & Matthes, E. (2019). Impact of anonymization on vehicle detector performance. 2019 Second International Conference on Artificial Intelligence for Industries (AI4I). https://doi.org/10.1109/ai4i46381.2019.00016) . Regarding Claim 16, Mukherjee teaches the method of claim 12, wherein the deception engine is configured to Mukherjee does not explicitly teach; However, Schnabel teaches: compare a performance of the raw data to a performance of the processed data received from the artificial intelligence engine (Schnabel, page 32, Section IV, discloses the training is conducted on both the original training set and different anonymized versions using the methods shown in III-B (object detection with CNNs) (abstract); Page 33, Section IV A & B, Schnabel discloses comparing Average Precision (AP) values from the original validation set with those from the anonymized validation set.) It would have been obvious to a person of ordinary skills in the art before the effective filing date to modify Mukherjee’s system to incorporate Schnabel's method of evaluating model accuracy on the original vs anonymized data to ensure that the obfuscation does not degrade the accuracy of the machine learning inference, provide quantitative evidence of compliance with privacy regulations without sacrificing model performance. One would be motivated to make such a modification on Mukherjee's system to verify that privacy measures do not reduce the utility of machine-learning inferences to achieve the predictable and desired result. Regarding Claim 19, Mukherjee teaches the method of claim 17, Mukherjee also discloses obfuscated fundamental metadata (para 22) Mukherjee does not explicitly teach; However, Schnabel teaches: verifying the obfuscated fundamental data by comparing a performance of the obfuscated fundamental metadata with a performance of the raw data (Schnabel, page 32, Section IV, discloses the training is conducted on both the original training set and different anonymized versions using the methods shown in III-B (object detection with CNNs) (abstract); Page 33, Section IV A & B, Schnabel discloses comparing Average Precision (AP) values from the original validation set with those from the anonymized validation set producing a validation result (Section V)) It would have been obvious to a person of ordinary skills in the art before the effective filing date to modify Mukherjee’s system to incorporate Schnabel's method of evaluating model accuracy on the original vs anonymized data to ensure that the obfuscation does not degrade the accuracy of the machine learning inference, provide quantitative evidence of compliance with privacy regulations without sacrificing model performance. One would be motivated to make such a modification on Mukherjee's system to verify that privacy measures do not reduce the utility of machine-learning inferences to achieve the predictable and desired result. Conclusion 07-39 AIA THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a). A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action. Any inquiry concerning this communication or earlier communications from the examiner should be directed to AMIT KHADKA whose telephone number is (703)756-1440. The examiner can normally be reached Monday - Friday, 8:00 am - 5:00 pm. 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, Jeffrey L. Nickerson can be reached at (469) 295-9235. 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. /AMIT KHADKA/Examiner, Art Unit 2432 /Jeffrey Nickerson/Supervisory Patent Examiner, Art Unit 2432 Application/Control Number: 17/815,876 Page 2 Art Unit: 2432 Application/Control Number: 17/815,876 Page 3 Art Unit: 2432 Application/Control Number: 17/815,876 Page 4 Art Unit: 2432 Application/Control Number: 17/815,876 Page 5 Art Unit: 2432 Application/Control Number: 17/815,876 Page 6 Art Unit: 2432 Application/Control Number: 17/815,876 Page 7 Art Unit: 2432 Application/Control Number: 17/815,876 Page 8 Art Unit: 2432 Application/Control Number: 17/815,876 Page 9 Art Unit: 2432 Application/Control Number: 17/815,876 Page 10 Art Unit: 2432 Application/Control Number: 17/815,876 Page 11 Art Unit: 2432 Application/Control Number: 17/815,876 Page 12 Art Unit: 2432 Application/Control Number: 17/815,876 Page 13 Art Unit: 2432 Application/Control Number: 17/815,876 Page 14 Art Unit: 2432 Application/Control Number: 17/815,876 Page 15 Art Unit: 2432 Application/Control Number: 17/815,876 Page 16 Art Unit: 2432 Application/Control Number: 17/815,876 Page 17 Art Unit: 2432 Application/Control Number: 17/815,876 Page 18 Art Unit: 2432 Application/Control Number: 17/815,876 Page 19 Art Unit: 2432 Application/Control Number: 17/815,876 Page 20 Art Unit: 2432 Application/Control Number: 17/815,876 Page 22 Art Unit: 2432 Application/Control Number: 17/815,876 Page 23 Art Unit: 2432
Prosecution Insights