DETAILED ACTION
This Office Action is in response to the most recent papers filed 6/22/2026.
Response to Arguments
Applicant’s arguments filed 6/22/2026 have been fully considered, but are moot as they focus on newly amended subject matter which is addressed with the new grounds of rejection necessitated by the amendments, as presented below.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 2015/0319072 (Chakrabarti) in view of US 10,524,141 (Guven) and US 2023/0214312 (Toal).
With regard to claim 1, Chakrabarti discloses method, the method comprising: at a test system:
performing, using test configuration information, a plurality of fuzz testing sessions involving one or more systems under test (SUT), wherein at least some of the plurality of fuzz testing sessions include different test traffic parameters and/or SUT configurations than at least one of the plurality of fuzz testing sessions (Chakrabarti: Abstract, Paragraphs [0029] and [0047, and Figures 7-9. Chakrabarti discloses fuzz testing, where at least test traffic can be provided for different testing sessions.);
obtaining fuzz test data from one or more sources, wherein the fuzz test data includes test traffic data and SUT performance data associated with the plurality of fuzz testing sessions (Chakrabarti: Paragraph [0103] and Figure 6.);
Chakrabarti fails to disclose, but Guven teaches that the method is for training a machine learning model for indicating a stress state value associated with a system under test (SUT); training, using the fuzz test data and one or more machine learning algorithms, a machine learning model for receiving as input traffic data associated with test traffic or live traffic involving a respective SUT and SUT performance data associated with the test traffic or live traffic and providing as output a stress state value indicating the likelihood of the respective SUT crashing or failing; and storing, in a machine learning model data store, the trained machine learning model for subsequent use by the test system or a SUT analyzer (Guven: Column 11, line 34 to Column 12, line 13. Guven provides for the training of machine learning models based on previous data of the nodes, where the model can use inputs to predict a likelihood of node failure. When this teaching is applied to fuzz testing, which serves to provide many different conditions to determine possible failures (Chakrabarti: Paragraph [0026]), such tests would serve as the previous data to train the models, with further input enabling additional inputs, whether test or live inputs, and provide a likelihood of node failure.).
Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to modify Chakrabarti to utilize the fuzz test data of Chakrabarti as training data for a machine learning model, such as in Guven, to improve the effectiveness of such fuzz testing, allowing the testing to be used to predict possible system failures beyond the limitations of the actual fuzz testing.
Chakrabarti fails to teach, but Toal teaches wherein the fuzz test data used in the training includes performance features that indicate that application programming interface (API) responses of the SUT are slower than the API responses were at initialization of the fuzz testing sessions of the SUT (Toal: Paragraphs [0039] and [0079]. In Toal, API success criteria can include response times and latency, where when these are too high (slower than a properly running API), this would indicate a failing API. Meanwhile, Guven, as applied above, teaches the learning of inputs that causes particular node behavior, such as node failures (Guven: Column 11, lines 40-62), where API responses becoming slower would be indicative of node failure, and thus would be part of the training data of Chakrabarti in view of Guven and Toal.).
Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to have the data used for training include features that indicate API responses are slower than at initialization, as in Toal, to predict the likelihood of node failure (Guven: Column 11, lines 40-62) using parameters that were known to indicate node failure (Toal: Paragraphs [0039] and [0079]).
With regard to claim 2, Chakrabarti in view of Guven fails to teach, but knowledge possessed by one of ordinary skill in the art at the time of filing teaches wherein the one or more machine learning algorithms includes an artificial neural network, a feedforward neural network, a recurrent neural network, or a convolutional neural network (More specifically, Official Notice is taken that each of these listed types of machine learning algorithms were well-known to one of ordinary skill in the art.). Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to utilize at least one of these well-known types of algorithms to realize the well-known benefits of such. For instance, Artificial Neural Networks were known to efficiently learn patterns from data and improve performance over time, with at least this type of algorithm working well with learning patterns from the fuzz testing and making predictions beyond the actual tests.
With regard to claim 3, Chakrabarti discloses wherein the fuzz test data used in training includes test results, wherein the test results include a final binary result indicating pass or failure for a respective SUT at the end of a respective fuzz testing session; and a final operator-provided value or metric for a respective SUT at the end of a respective fuzz testing session (Chakrabarti: Paragraphs [0024] and [0067]. The fuzz testing indicates failures of the device under test.).
Chakrabarti fails to teach, but knowledge possessed by one of ordinary skill in the art at the time of filing teaches that the value or metric is on a predetermined scale indicating a likelihood or nearness to failure (More specifically, Official Notice is taken that the providing of expected output by a human operator for training machine learning models was well-known to one of ordinary skill in the art at the time of filing, where Guven teaches the output of a likelihood of node failure (Guven: Column 11, lines 34 to 62).). Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to have a human operator, such as that of Chakrabarti, provide a likelihood of failure in a manner that would be expected for an output of the system (predetermined scale) to improve the training of the machine learning model to ensure that examples are provided of the actual expected output for at least some of the tests.
With regard to claim 4, the instant claim is within the scope of claim 3, and is rejected for similar reasons.
With regard to claim 5, Chakrabarti teaches wherein the fuzz test data used in training is correlated using timestamps (Chakrabarti: Paragraph [0025]).
With regard to claim 7, Chakrabarti in view of Guven teaches at the SUT analyzer: receiving, via the test system or the machine learning model data store, the trained machine learning model; receiving traffic data and SUT performance data associated with network traffic involving a first SUT; generating, using the traffic data and the SUT performance data as input to the trained machine learning model, a stress state value associated with the first SUT; and providing the stress state value to a display, a user, or another entity (Chakrabarti: Paragraph [0035] and Guven: Column 11, line 34 to Column 12, line 13. In Chakrabarti, the system may be controllable by a human operator, such that the human is presented different GUIs for controlling the tests and/or reviewing the results. Meanwhile, in Guven, the machine learning model, trained based on previous data, can be used for making predications based on additional data, which when such is under control of a human operator, would present those results to the human as well.).
With regard to claim 8, Chakrabarti in view of Guven teaches at the test system: receiving traffic data and SUT performance data associated with an on-going or completed first fuzz testing session involving a first SUT; generating, using the traffic data and the SUT performance data as input to the trained machine learning model, a stress state value associated with the first SUT; and providing the stress state value to a display, a user, or another entity (Chakrabarti: Paragraph [0035] and Guven: Column 11, line 34 to Column 12, line 13. In Chakrabarti, the system may be controllable by a human operator, such that the human is presented different GUIs for controlling the tests and/or reviewing the results. Meanwhile, in Guven, the machine learning model, trained based on previous data, can be used for making predications based on additional data, which when such is under control of a human operator, would present those results to the human as well.).
With regard to claim 9, Chakrabarti in view of Guven teaches wherein the traffic data includes copies of network traffic, log data, or traffic metrics and at least some of the traffic data is obtained from the test system, a fuzz testing module, a traffic generator, a monitoring agent, a network probe, a network tap, one or more data repositories, or the SUT (Chakrabarti: Figure 6. The at least the fuzz testing module would provide traffic metrics (data about the traffic).); and wherein the SUT performance data includes performance or health statistics or metrics, SUT state information, error information, or failure information and at least some of the SUT performance data is obtained from the test system, the one or more data repositories, or the SUT (Chakrabarti: Paragraph [0044]. At least failure information would be received from, for example, the test system.).
With regard to claims 10-14 and 16-18, the instant claims are similar to claims 1-5 and 7-9, and are rejected for similar reasons.
With regard to claims 19-20, the instant claims are similar to claims 1-2, and are rejected for similar reasons.
Claim Rejections - 35 USC § 103
Claim(s) 6 and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over US Chakrabarti in view of Guven and Toal, and further in view of Mismar et al. in “Unsupervised Learning in Next-Generation Networks: Real-Time Performance Self-Diagnosis,” Published June 27, 2021 (Mismar).
With regard to claim 6, Chakrabarti fails to teach, but Mismar teaches the training of the machine learning model utilizes an unsupervised learning technique (Mismar: Mismar teaches that unsupervised learning techniques were well-known to one of ordinary skill in the art at the time of filing.). Accordingly, it would have been obvious to one of ordinary skill in the art at the time of filing to utilize unsupervised learning techniques to reduce the processing of training data versus supervised learning techniques. Further, in general, there would be unsupervised or supervised learning techniques, where one of ordinary skill in the art would have been motivated to pursue each of these two options that are within their technical grasp, where each of the two techniques provide known advantages and disadvantages.
With regard to claim 15, the instant claim is similar to claim 6, and is rejected for similar reasons.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any 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 SCOTT B CHRISTENSEN whose telephone number is (571)270-1144. The examiner can normally be reached Monday through Friday, 6AM to 2PM.
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, John Follansbee can be reached at (571) 272-3964. 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.
SCOTT B. CHRISTENSEN
Examiner
Art Unit 2444
/SCOTT B CHRISTENSEN/Primary Examiner, Art Unit 2444
Prosecution Insights