Patents by Inventor Ariel Faigon
Ariel Faigon has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20230344841Abstract: The technology relates to machine responses to anomalies detected using machine learning based anomaly detection. In particular, to receiving evaluations of production events, prepared using activity models constructed on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, to responding to detected anomalies in near real-time streams of security-related events of tenants, the anomalies detected by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant.Type: ApplicationFiled: July 5, 2023Publication date: October 26, 2023Inventors: Jeevan Tambuluri, Ravi Ithal, Steve Malmskog, Abhay Kulkarni, Ariel Faigon, Krishna Narayanaswamy
-
Patent number: 11743275Abstract: The technology relates to machine responses to anomalies detected using machine learning based anomaly detection. In particular, to receiving evaluations of production events, prepared using activity models constructed on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, to responding to detected anomalies in near real-time streams of security-related events of tenants, the anomalies detected by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant.Type: GrantFiled: May 27, 2021Date of Patent: August 29, 2023Assignee: Netskope, Inc.Inventors: Jeevan Tambuluri, Ravi Ithal, Steve Malmskog, Abhay Kulkarni, Ariel Faigon, Krishna Narayanaswamy
-
Publication number: 20210288983Abstract: The technology relates to machine responses to anomalies detected using machine learning based anomaly detection. In particular, to receiving evaluations of production events, prepared using activity models constructed on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, to responding to detected anomalies in near real-time streams of security-related events of tenants, the anomalies detected by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant.Type: ApplicationFiled: May 27, 2021Publication date: September 16, 2021Applicant: Netskope, Inc.Inventors: Jeevan TAMBULURI, Ravi ITHAL, Steve MALMSKOG, Abhay KULKARNI, Ariel FAIGON, Krishna NARAYANASWAMY
-
Patent number: 11025653Abstract: The technology disclosed relates to machine learning based anomaly detection. In particular, it relates to constructing activity models on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, it relates to detecting anomalies in near real-time streams of security-related events of one or more tenants by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant. It further includes determining an anomaly score for a production event based on calculated likelihood coefficients of categorized feature-value pairs and a prevalencist probability value of the production event comprising the coded features-value pairs.Type: GrantFiled: April 19, 2019Date of Patent: June 1, 2021Assignee: Netskope, Inc.Inventors: Ariel Faigon, Krishna Narayanaswamy, Jeevan Tambuluri, Ravi Ithal, Steve Malmskog, Abhay Kulkarni
-
Publication number: 20190245876Abstract: The technology disclosed relates to machine learning based anomaly detection. In particular, it relates to constructing activity models on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, it relates to detecting anomalies in near real-time streams of security-related events of one or more tenants by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant. It further includes determining an anomaly score for a production event based on calculated likelihood coefficients of categorized feature-value pairs and a prevalencist probability value of the production event comprising the coded features-value pairs.Type: ApplicationFiled: April 19, 2019Publication date: August 8, 2019Applicant: Netskope, Inc.Inventors: Ariel FAIGON, Krishna NARAYANASWAMY, Jeevan TAMBULURI, Ravi ITHAL, Steve MALMSKOG, Abhay KULKARNI
-
Patent number: 10270788Abstract: The technology disclosed relates to machine learning based anomaly detection. In particular, it relates to constructing activity models on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, it relates to detecting anomalies in near real-time streams of security-related events of one or more tenants by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant. It further includes determining an anomaly score for a production event based on calculated likelihood coefficients of categorized feature-value pairs and a prevalencist probability value of the production event comprising the coded features-value pairs.Type: GrantFiled: September 2, 2016Date of Patent: April 23, 2019Assignee: Netskope, Inc.Inventors: Ariel Faigon, Krishna Narayanaswamy, Jeevan Tambuluri, Ravi Ithal, Steve Malmskog, Abhay Kulkarni
-
Publication number: 20170353477Abstract: The technology disclosed relates to machine learning based anomaly detection. In particular, it relates to constructing activity models on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, it relates to detecting anomalies in near real-time streams of security-related events of one or more tenants by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant. It further includes determining an anomaly score for a production event based on calculated likelihood coefficients of categorized feature-value pairs and a prevalencist probability value of the production event comprising the coded features-value pairs.Type: ApplicationFiled: September 2, 2016Publication date: December 7, 2017Applicant: Netskope, Inc.Inventors: Ariel FAIGON, Krishna NARAYANASWAMY, Jeevan TAMBULURI, Ravi ITHAL, Steve MALMSKOG, Abhay KULKARNI
-
Patent number: 8583778Abstract: Techniques are provided through which “suspicious” websites may be identified automatically. A suspicious website is one that is associated with many changes or an inconsistent number of changes in web registry information over time. Registry information is received when changes to the registry information occur. The registry information is referred to as a transaction. A transaction is comprised of a plurality of values that each correspond to a characteristic. A characteristic is a property of a website, such as the website's contact information. A count associated with a particular characteristic-value pair is updated each time the particular value is identified in a transaction. A high count indicates that the website associated with the particular value is associated with a lot of changes. Therefore, a website associated with a high count is suspicious. Other factors that may be used for identifying a “suspicious” website include how often and how much the count changes.Type: GrantFiled: April 26, 2006Date of Patent: November 12, 2013Assignee: Yahoo! Inc.Inventors: Ariel Faigon, Timothy M. Converse, Priyank S. Garg