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
This Office Action is in response to the application 19/088,247 filed on 03/24/2025.
Claims 1-20 have been examined and are pending in this application.
Examiner’s notes
Claims 8-13 recites “One or more computer-readable media;” The specification explicitly defines as to what type of computer readable storage medium is claimed. In [par. 402; lines 7-9], the specification discloses “The terms computer-readable media or computer-readable storage media do not include signals and carrier waves. In addition, the terms computer-readable media or computer-readable storage media do not include communication ports (e.g., 1270) or communication media.” Therefore, the claims are not directed to non-statutory subject matter. However, The Examiner respectfully suggests that the claims be amended to either “A non-transitory computer readable storage medium” or “a computer readable storage device”.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(B) CONCLUSION. —The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b), as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, or for pre-AIA the applicant regards as the invention.
Regarding claims 1, 8 and 14, claims 1, 8 and 14 recite “a first machine-learning (ML) tool, comprising: at training time: …….” which is unclear. It is not clear how a machine learning(ML) tool comprising at training time:
Regarding claims 1, 8 and 14, claims 1, 8 and 14 recite “(a) training a second machine-learning (ML) tool to learn language or domain knowledge…….” which is unclear. It is not clear to learn language or domain knowledge of what.
Regarding claims 1, 8 and 14, claims 1, 8 and 14 recite “(b) training the second ML tool to classify inputted data based on sensitivity and (c) inputting, to the second ML tool, groups of input data based on output generated by the first ML tool …….” which is unclear. It is not clear the inputted data of operation (b) and the input data based on output of first machine learning tool is same or different.
Regarding claims 1, 8 and 14, claims 1, 8 and 14 recite “(b) training the second ML tool to classify inputted data based on sensitivity and (c) inputting, to the second ML tool, groups of input data based on output generated by the first ML tool …….” which is unclear. It is not clear which operations is first, the operation (b) or the operation (c) when the inputter data same.
Regarding claims 2-7, 9-13 and 15-20; claims 2-7, 9-13 and 15-20 are dependent on claims 1, 8 and 14, and are analyzed and rejected accordingly.
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 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 of this title, 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 1-3, 8-10 and 14-16 are rejected under 35 U.S.C. 103 as being unpatentable over Ganguly (US 2022/0366074), in view of Narayanaswamy (US 2022/0294831).
Regarding claim 1, Ganguly discloses a computer-implemented method of monitoring output of a first machine-learning (ML) tool (Ganguly par. 0037; The gender model may output a predicted gender as well as a confidence value describing how confident the gender model is in its prediction. As this is repeated with different models, a list of vulnerable sensitive types can be created and maintained. See also par. 0055), comprising:
at training time:
(a) training a second machine-learning (ML) tool to learn language or domain knowledge (Ganguly par. 0055; For example, the overlap cluster used to train a machine learning model to accomplish the goal task with an accuracy); and
(b) training the second ML tool to classify inputted data based on sensitivity (Ganguly par. 0037, 0055 and 0065; For example, the overlap cluster may be used to train a machine learning model to accomplish the goal task with an accuracy. At the same time, that overlap cluster used by another machine learning model to identify sensitive features with an accuracy making the overlap cluster relevant to sensitive data. A list of sensitive data types based solely on available models is not necessarily exhaustive; other possible ways of identifying vulnerable sensitive data types without preexisting models are also considered, such as attempting to train a new machine learning model to extract sensitive data of a given type (utilizing a received data space including annotations describing the given type as training data)); and
at inference time:
(c) inputting, to the second ML tool, groups of input data based on output generated by the first ML tool (Ganguly par. 0038 and 0066; In some instances, the input data space may include annotations describing sensitive information, in which case operation include attempting to train a machine learning model based upon the data space and determining an accuracy of the trained model. Method 300 further comprises attempting to extract sensitive data from the data space via one or more pretrained models at operation 306. Operation 306 may include, for example, inputting the data space received at operation 302 to an existing machine learning model and receiving an output from the model. For example, a system performing method 300 may have selected, at operation 304, “age” as a sensitive data type. In such an example, the system may, at operation 306, input the data space into a model that has been trained to extract age information and receive an output from the model);
(d) receiving, from the second ML tool, respective classifications of each group of input data (Ganguly par. 0065; In some instances, the list may be based on a number of available machine learning models. As each type of sensitive data can be checked for vulnerability based upon a model trained to extract that type of sensitive data (as discussed in further detail below with reference to operation 306), a list of available machine learning models may constitute a list of sensitive data types that the system performing method 300 can be expected to evaluate. For example, a system performing method 300 may have access to three trained models; one to determine age, one to determine gender, and one to determine an address);
for a first group having a classification indicating that the respective input data is sensitive:
(e) flagging, discarding, or redacting at least a portion of the sensitive input data (Ganguly par. 0067; Method 300 further comprises comparing the confidence to a confidence threshold at operation 308. The confidence threshold may be preset. For example, in some instances, operation 308 may include checking if the model confidence is above 50%. If the confidence is above the confidence threshold (308 “Yes”), method 300 further comprises flagging the selected sensitive data type as “vulnerable” at operation 310. See also par. 0070).
Ganguly teaches, classifications of each group of input data and flagging sensitive data (Ganguly par. 0065 and 0067). However, Ganguly does not explicitly disclose for a second group having a classification indicating that the respective input data is not sensitive: (f) forwarding the second group to a destination.
However, in an analogous art, Narayanaswamy teaches for a second group having a classification indicating that the respective input data is not sensitive: (f) forwarding the second group to a destination (Narayanaswamy par. 0140; The device comprises a local metadata store maintained at the endpoint and configured to periodically receive from a cloud-based metadata store sensitivity metadata previously generated to classify documents as sensitive or non-sensitive based on deep inspection of the documents; a local anchor pattern scanner running on the endpoint and configured to preliminarily classify the documents as sensitive or non-sensitive based on anchor pattern check, to send the documents that scored positive on the anchor pattern check to a cloud-based sensitivity scanner that confirmatory classifies the documents as sensitive or non-sensitive based on deep inspection).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the teachings of Narayanaswamy with the method and system of Ganguly, wherein a second group having a classification indicating that the respective input data is not sensitive forwarding the second group to a destination to provide users with a means for classifying the document as sensitive or non-sensitive based on deep inspection (Narayanaswamy par. 0137).
Regarding claim 2, Ganguly and Narayanaswamy disclose the computer-implemented method of claim 1,
Ganguly further discloses wherein act (b) is performed using training data labeled based on the sensitivity (Ganguly par. 0065; other possible ways of identifying vulnerable sensitive data types without preexisting models are also considered, such as attempting to train a new machine learning model to extract sensitive data of a given type (utilizing a received data space including annotations describing the given type as training data)).
Regarding claim 3, Ganguly and Narayanaswamy disclose the computer-implemented method of claim 1,
Ganguly further discloses further comprising: using the first group for fine-tuning the first ML tool (Ganguly par. 0038; In some instances, the input data space may include annotations describing sensitive information, in which case operation include attempting to train a machine learning model based upon the data space and determining an accuracy of the trained model).
Regarding claims 8-10; claims 8-10 are directed to a computer-readable media associated with the method claimed in claims 1-3 respectively. Claims 8-10 are similar in scope to claims 1-3 respectively, and are therefore rejected under similar rationale respectively.
Regarding claims 14-16; claims 14-16 are directed to a system associated with the method claimed in claims 1-3 respectively. Claims 14-16 are similar in scope to claims 1-3 respectively, and are therefore rejected under similar rationale respectively.
Claims 4-7, 11-13 and 17-20 are rejected under 35 U.S.C. 103 as being unpatentable over Ganguly (US 2022/0366074), in view of Narayanaswamy (US 2022/0294831) and further in view of Khanna (US 2024/0176674).
Regarding claim 4, Ganguly and Narayanaswamy disclose the computer-implemented method of claim 1,
Ganguly and Narayanaswamy failed to disclose but Khanna discloses wherein the first ML tool or the second ML tool is part of a microservice in a network of microservices configured as a copilot (Ganguly par. 0057; An example embodiment of an architecture for facilitating various machine learning functions of the online concierge system 140 using microservices is described in further detail below with respect to FIG. 3. See also par. 0059).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the teachings of Khanna with the method and system of Ganguly and Narayanaswami, wherein the first ML tool or the second ML tool is part of a microservice in a network of microservices configured as a copilot to provide useful, accurate, and cost-efficient results while effectively managing latency (Khanna par. 0001).
Regarding claim 5, Ganguly, Narayanaswamy and Khanna disclose the computer-implemented method of claim 4,
Khanna further discloses wherein the first ML tool is part of a core microservice of the copilot (Ganguly par. 0057; An example embodiment of an architecture for facilitating various machine learning functions of the online concierge system 140 using microservices is described in further detail below with respect to FIG. 3. See also par. 0052 and 0059).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the teachings of Khanna with the method and system of Ganguly and Narayanaswami, wherein the first ML tool or the second ML tool is part of a microservice in a network of microservices configured as a copilot to provide useful, accurate, and cost-efficient results while effectively managing latency (Khanna par. 0001).
Regarding claim 6, Ganguly, Narayanaswamy and Khanna disclose the computer-implemented method of claim 4,
Khanna further discloses wherein the first ML tool is part of a data producer or a retrieval microservice of the copilot (Khanna par. 0014; he tuning techniques are described herein primarily in the context of an online concierge system 140. However, the online concierge system 140 represents just one example of an online system in which machine learning microservices may be utilized to perform various system functions and tuned according to the embodiments of this disclosure. See also par. 0052).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the teachings of Khanna with the method and system of Ganguly and Narayanaswami, wherein the first ML tool or the second ML tool is part of a microservice in a network of microservices configured as a copilot to provide useful, accurate, and cost-efficient results while effectively managing latency (Khanna par. 0001).
Regarding claim 7, Ganguly, Narayanaswamy and Khanna disclose the computer-implemented method of claim 4,
Narayanaswamy further discloses second group having a classification indicating that the respective input data is not sensitive and forwarded at act (f) (Narayanaswamy par. 0140; The device comprises a local metadata store maintained at the endpoint and configured to periodically receive from a cloud-based metadata store sensitivity metadata previously generated to classify documents as sensitive or non-sensitive based on deep inspection of the documents; a local anchor pattern scanner running on the endpoint and configured to preliminarily classify the documents as sensitive or non-sensitive based on anchor pattern check, to send the documents that scored positive on the anchor pattern check to a cloud-based sensitivity scanner that confirmatory classifies the documents as sensitive or non-sensitive based on deep inspection).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the teachings of Narayanaswamy with the method and system of Ganguly, wherein a second group having a classification indicating that the respective input data is not sensitive forwarding the second group to a destination to provide users with a means for classifying the document as sensitive or non-sensitive based on deep inspection (Narayanaswamy par. 0137).
Khanna further discloses wherein the second group is forwarded at act (f) toward a client of the copilot (Khanna par. 0057; In this architecture, each microservice may facilitate an independent machine learning algorithm for training and/or inferences. For example, microservices may be utilized for features such as ranking search results for items in the online concierge system 140, predicting item availability, predicting delivery times associated with a placed order, determining promotions to serve to a user, assigning pickers to orders, or other tasks. An example embodiment of an architecture for facilitating various machine learning functions of the online concierge system 140 using microservices ).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the teachings of Khanna with the method and system of Ganguly and Narayanaswami, wherein the first ML tool or the second ML tool is part of a microservice in a network of microservices configured as a copilot to provide useful, accurate, and cost-efficient results while effectively managing latency (Khanna par. 0001).
Regarding claims 11-13; claims 11-13 are directed to a computer-readable media associated with the method claimed in claims 4-6 respectively. Claims 8-13 are similar in scope to claims 4-6 respectively, and are therefore rejected under similar rationale respectively.
Regarding claims 17-20; claims 17-20 are directed to a system associated with the method claimed in claims 4-7 respectively. Claims 17-20 are similar in scope to claims 4-7 respectively, and are therefore rejected under similar rationale respectively.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SANCHIT K SARKER whose telephone number is (571)270-7907. The examiner can normally be reached M-F 8:30 AM-5:30 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, FARID HOMAYOUNMEHR can be reached at 571-272-3739. 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.
/SANCHIT K SARKER/Primary Examiner, Art Unit 2495
Prosecution Insights