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 .
DETAILED ACTION
1. Claims 1 - 20 are pending. Claims 1, 8, 15 are independent. File date on 7-31-2025.
Claim Rejections - 35 USC § 103
2. 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.
3. Claims 1, 3, 4, 6, 8, 10, 11, 13, 15, 17, 18, 20 are rejected under 35 U.S.C. 103 as being unpatentable over Khan et al. (US PGPUB No. 20190004737) in view of Kumar et al. (US Patent No. 10,915,417) and further in view of Boven et al. (US PGPUB No. 20210034581).
Regarding Claims 1, 8, 15, Khan discloses a method and a system and a non-transitory computer-readable medium storing instructions comprising:
a) identifying a transformed dataset stored in a storage system and one or more transformations applied to generate the transformed dataset, wherein the fingerprint indicates a performance of one or more artificial intelligence (Al) model executions that used the transformed dataset as input; (Khan ¶ 062: flow for processing a multi-object transformation command in accordance with certain embodiments. At 502, host 101 issues a multi-object transformation command (e.g., an object definition command specifying a transform involving multiple objects) with a transformation type (i.e., an indication of which transform operation should be performed) and corresponding object IDs (i.e., identifications of each object that is an input into the transformation; ¶ 055, ll 1-13: storage device applies transformation specified in object definition model command; results stored in memory; ¶ 018: discover object transformation capabilities, issue commands to storage devices specifying objects and transformations to apply to set of objects; perform layered processing using storage devices for analytics and processing utilizing machine learning applications)
Khan does not explicitly disclose for a) generating a fingerprint, wherein the fingerprint indicates a performance of one or more model executions that used the transformed dataset as input.
However, Kumar discloses for a) generating a fingerprint, wherein the fingerprint indicates a performance of one or more artificial intelligence (Al) model executions that used the transformed dataset as input. (Kumar col 2: a system for auditing transformed data in accordance with an embodiment. Specifically, FIG. 1 depicts an example 100 of an audit system 122 auditing transformed data, whereby a client device 102 submits, to a data store manager 106 via one or more application programming interfaces (APIs), such as API 104, commands to perform operations, which may be mutating operations, on data of a customer of a computing resource service provider. A first data storage system 108 obtains the commands from the data store manager 106 and generates a first log 112 of the operations. A second data storage system 110 further obtains the list of operations from the first log 112 and generates a second log 114 of the operations.; col 5: the first data storage system 108 and second data storage system 110 may analyze a set of entries of the first log 112 and a corresponding set of entries of the second log 114 respectively, and generate representations (e.g., checksums or other “fingerprints” of the data) of each of the sets of entries, which may be referred to in the present disclosure as “audit information.” ... the first data storage system 108 and second data storage system 110 may audit the log entries by applying various transformations to aspects of the log entries to produce audit information that can be compared to validate the log entries.; (fingerprint information))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Khan for a) generating auditing information based on the fingerprint, the auditing information describing the transformed dataset and the one or more transformations applied as taught by Kumar. One of ordinary skill in the art would have been motivated to employ the teachings of Kumar for the benefits achieved from the flexibility of a system that enables the utilization of multiple techniques such as generation of a fingerprint associated with audit information. (Kumar col 2; col 5)
Khan does not explicitly disclose for b) selecting the transformed dataset for reuse based on a determination, that the transformed dataset is suitable for use with the subsequent AI model execution.
However, Boven discloses wherein for b) selecting, for a subsequent AI model execution, the transformed dataset for reuse based on a determination that the transformed dataset is suitable for use with the subsequent AI model execution. (Boven ¶ 006: executing the created data science model by using the indicated portion of the transformed asset-related data stored in the second data store as inputs to the data science model, and (j) causing at least of portion of an output of the data science model to be rendered according to the application configuration parameters.; (previously store transformed data utilized as input a model))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Khan for for b) selecting the transformed dataset for reuse based on a determination, that the transformed dataset is suitable for use with the subsequent AI model execution as taught by Boven. One of ordinary skill in the art would have been motivated to employ the teachings of Boven for the benefits achieved from the flexibility of a system that enables the input of previously stored transformed datasets thereby reducing processing time and usage of resources. (Boven ¶ 006)
Furthermore, for Claim 8, Khan discloses wherein one or more processors; and memory storing instructions that, when executed by the one or more processors, cause the system to perform operations. (Khan ¶ 029: A system memory device 107 may store any suitable data, such as data used by CPU 102 to provide the functionality of computer system 100. For example, data associated with programs that are executed by or files accessed by cores 114 may be stored in system memory device 107. Thus, a system memory device 107 may include a system memory (sometimes referred to as a main memory) that stores data and/or sequences of instructions that are executed or otherwise used by the cores 114.)
Furthermore, for Claim 15, Khan discloses wherein a non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause a system to perform operations. (Khan ¶ 029: A system memory device 107 may store any suitable data, such as data used by CPU 102 to provide the functionality of computer system 100. For example, data associated with programs that are executed by or files accessed by cores 114 may be stored in system memory device 107. Thus, a system memory device 107 may include a system memory (sometimes referred to as a main memory) that stores data and/or sequences of instructions that are executed or otherwise used by the cores 114.)
Regarding Claims 3, 10, 17, Khan discloses the method of claim 1 and the system of claim 8 and the computer-readable medium of claim 15, wherein the fingerprint further identifies one or more characteristics of input data used during the one or more AI model executions. (Khan ¶ 050: in response to the “get features” command, the storage device 106 may respond to the host with the object transformation capabilities of the storage device 106. The object transformation capabilities may include a list of the object types supported by the SSD and what types of transformations may be performed by the storage device 106 for each object type supported by the SSD.; (capabilities input data))
Regarding Claims 4, 11, 18, Khan discloses the method of claim 1 and the system of claim 8 and the computer-readable medium of claim 15, further comprising:
Khan does not explicitly disclose for a) generating auditing information based on the fingerprint, the auditing information describing the transformed dataset and the one or more transformations applied, and for b) validating one or more model executions using the auditing information.
However, Kumar discloses:
a) generating auditing information based on the fingerprint, the auditing information describing the transformed dataset and the one or more transformations applied; and b) validating one or more AI model executions using the auditing information. (Kumar col 2: a system for auditing transformed data in accordance with an embodiment. Specifically, FIG. 1 depicts an example 100 of an audit system 122 auditing transformed data, whereby a client device 102 submits, to a data store manager 106 via one or more application programming interfaces (APIs), such as API 104, commands to perform operations, which may be mutating operations, on data of a customer of a computing resource service provider. A first data storage system 108 obtains the commands from the data store manager 106 and generates a first log 112 of the operations. A second data storage system 110 further obtains the list of operations from the first log 112 and generates a second log 114 of the operations.; col 5: the first data storage system 108 and second data storage system 110 may analyze a set of entries of the first log 112 and a corresponding set of entries of the second log 114 respectively, and generate representations (e.g., checksums or other “fingerprints” of the data) of each of the sets of entries, which may be referred to in the present disclosure as “audit information.” ... the first data storage system 108 and second data storage system 110 may audit the log entries by applying various transformations to aspects of the log entries to produce audit information that can be compared to validate the log entries.; (fingerprint information))
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Khan for a) generating auditing information based on the fingerprint, the auditing information describing the transformed dataset and the one or more transformations applied, and for b) validating one or more model executions using the auditing information as taught by Kumar. One of ordinary skill in the art would have been motivated to employ the teachings of Kumar for the benefits achieved from the flexibility of a system that enables the utilization of multiple techniques such as generation of a fingerprint associated with audit information. (Kumar col 2; col 5)
Regarding Claims 6, 13, 20, Khan discloses the method of claim 1 and the system of claim 8 and the computer-readable medium of claim 15, including storage of transformed data sets. (Khan ¶ 055, ll 1-13: storage device applies transformation specified in object definition command; results stored in memory; results retrievable from storage device at any suitable time; (access transformed dataset without having to reapply transformation to obtain transformed dataset))
Khan does not explicitly disclose bypassing a re-performance of the one or more transformations in response to selecting the transformed dataset for reuse.
However, Boven discloses wherein further comprising bypassing a re-performance of the one or more transformations in response to selecting the transformed dataset for reuse. (Boven ¶ 006: executing the created data science model by using the indicated portion of the transformed asset-related data stored in the second data store as inputs to the data science model, and (j) causing at least of portion of an output of the data science model to be rendered according to the application configuration parameters.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Khan for bypassing a re-performance of the one or more transformations in response to selecting the transformed dataset for reuse as taught by Boven. One of ordinary skill in the art would have been motivated to employ the teachings of Boven for the benefits achieved from the flexibility of a system that enables the input of previously stored transformed datasets thereby reducing processing time and usage of resources. (Boven ¶ 006)
4. Claims 2, 9, 16 are rejected under 35 U.S.C. 103 as being unpatentable over Khan in view of Kumar, Boven and further in view of Grimm et al. (US PGPUB No. 20140131524).
Regarding Claims 2, 9, 16, Khan discloses the method of claim 1 and the system of claim 8 and the computer-readable medium of claim 15.
Khan does not explicitly disclose the transform related object is generated based on a hash of the transformed dataset and the one or more transformations.
However, Grimm discloses wherein the transform related object is generated based on a hash of the transformed dataset and the one or more transformations. (Grimm ¶ 064: Management System 142 may be programmed and/or configured to receive and store the Hash Data calculated by Office Segment 112 based on the Train Authority Data transformed by the Office Segment 112. The Management System 142 may be further programmed and/or configured to compare the Hash Data (e.g., CRC) calculated by the Office Segment 112 with the Hash Data calculated by the Management System 142 and determine whether a transformation error and/or inconsistency occurred. In a non-limiting exemplary implementation, when the Management System 142 determines that the calculated Hash Data for the Train Authority Data transformed by the Management System 142 does not match the calculated Hash Data for the Train Authority Data transformed by the Back Office System 108, the Management System 142 may be programmed and/or configured to execute at least one action.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Khan for the transform related object is generated based on a hash of the transformed dataset and the one or more transformations as taught by Grimm. One of ordinary skill in the art would have been motivated to employ the teachings of Boven for the benefits achieved from the flexibility of a system that enables the utilization of multiple techniques such as secure hash calculations in the processing of transformed datasets. (Grimm ¶ 064)
5. Claims 5, 12, 19 are rejected under 35 U.S.C. 103 as being unpatentable over Khan in view of Kumar and further in view of Martinez Iraola et al. (US PGPUB No. 20140131524, referred to as “MartinezIraola”).
Regarding Claims 5, 12, 19, Khan discloses the method of claim 1 and the system of claim 8 and the computer-readable medium of claim 15,
Khan does not explicitly disclose recommending the one or more transformations for use in generating another transformed dataset based on the fingerprint.
However, MartinezIraola discloses wherein further comprising recommending the one or more transformations for use in generating another transformed dataset based on the fingerprint. (MartinezIraola ¶ 004: a set of recommendations for transforming the input dataset, and transforming a text of the input dataset based on the set of recommendations)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Khan for recommending the one or more transformations for use in generating another transformed dataset based on the fingerprint as taught by MartinezIraola. One of ordinary skill in the art would have been motivated to employ the teachings of MartinezIraola for the benefits achieved from the flexibility of a system that enables the utilization of multiple techniques such as implementation of recommendations for selection of transformation types. (MartinezIraola ¶ 004)
6. Claims 7, 14 are rejected under 35 U.S.C. 103 as being unpatentable over Khan in view of Kumar and further in view of Walters et al. (US PGPUB No. 20200012657).
Regarding Claims 7, 14, Khan discloses the method of claim 1 and the system of claim 8.
Khan does not explicitly disclose selecting the transformed dataset comprises determining a similarity between the subsequent Al model execution and one or more prior AI model executions associated with the fingerprint.
However, Walters discloses wherein selecting the transformed dataset comprises determining a similarity between the subsequent Al model execution and one or more prior AI model executions associated with the fingerprint. (Walters ¶ 050: the similarity metric represents a statistical similarity between data-model output of a first dataset and a second dataset, consistent with disclosed embodiments. As another example, data classification module may classify a dataset as a related dataset based on determination that a similarity metric between a dataset and a previously classified dataset satisfies a criterion.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Khan for selecting the transformed dataset comprises determining a similarity between the subsequent Al model execution and one or more prior AI model executions associated with the fingerprint as taught by Walters. One of ordinary skill in the art would have been motivated to employ the teachings of Walters for the benefits achieved from the flexibility of a system that enables the utilization of multiple techniques such as a determination of similarity between different datasets in the processing of transformed datasets. (Walters ¶ 050)
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Kyung H Shin whose telephone number is (571)272-3920. The examiner can normally be reached M - F: 12pm - 8pm.
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, Joon H Hwang can be reached at 571-272-4036. 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.
/KYUNG H SHIN/ 9-27-2026Primary Examiner, Art Unit 2447
Prosecution Insights