Patents by Inventor Richa Singh
Richa Singh 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).
-
Patent number: 11763159Abstract: A neural network is configured to suppress an output of a mitigation node in a mitigation layer of the neural network. The neural network is pre-configured to recognize objects from inputs when operating using a processor and a memory. An actual input is sent to the neural network for object recognition, the actual input is an altered input. By suppressing the output of the mitigation node, the neural network is caused to avoid falsely recognizing an object from the actual input, where the altered input is configured to cause the neural network to falsely recognize the object from the actual input.Type: GrantFiled: January 29, 2018Date of Patent: September 19, 2023Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Gaurav Goswami, Sharathchandra Pankanti, Nalini K. Ratha, Richa Singh, Mayank Vatsa
-
Publication number: 20230013612Abstract: Described are a system, method, and computer program product for real-time payment gateway event monitoring. The method includes receiving input data from a merchant system including an application URL associated with a merchant control interface having a web socket programmed and/or configured to persistently listen for HTTP POST messages, and an event of a payment gateway system. The method also includes monitoring ongoing events of the payment gateway system associated with ongoing transaction messages communicated from point-of-sale (POS) devices to the payment gateway system and detecting the event in the ongoing events. The method further includes, in response to detection of the event, generating a HTTP POST message including event data of the event, and communicating the HTTP POST message to the application URL to cause the merchant control interface to display the event data in a chart for visual representation of the event data.Type: ApplicationFiled: September 19, 2022Publication date: January 19, 2023Inventors: Anurag Gupta, Jagadeesh Saravanan, Richa Singh
-
Patent number: 11475463Abstract: Described are a system, method, and computer program product for real-time payment gateway event monitoring. The method includes receiving input data from a merchant system including an application URL associated with a merchant control interface having a web socket programmed and/or configured to persistently listen for HTTP POST messages, and an event of a payment gateway system. The method also includes monitoring ongoing events of the payment gateway system associated with ongoing transaction messages communicated from point-of-sale (POS) devices to the payment gateway system and detecting the event in the ongoing events. The method further includes, in response to detection of the event, generating a HTTP POST message including event data of the event, and communicating the HTTP POST message to the application URL to cause the merchant control interface to display the event data in a chart for visual representation of the event data.Type: GrantFiled: August 14, 2020Date of Patent: October 18, 2022Assignee: Visa International Service AssociationInventors: Anurag Gupta, Jagadeesh Saravanan, Richa Singh
-
Publication number: 20210287567Abstract: Covers various features of the Quikik product, specifically a feature for skill testing against an AI that determines pace through course content, and a feature for requesting answers from instructors, possibly utilizing answers to previously-answered questions or software solutions to math problems.Type: ApplicationFiled: March 11, 2021Publication date: September 16, 2021Inventors: Ujjwal SINGH, Martha J. REVENAUGH, Richa SINGH, Manik KAKAR, Nancy Elizabeth MOORE
-
Patent number: 10944767Abstract: Mechanisms are provided for training a classifier to identify adversarial input data. A neural network processes original input data representing a plurality of non-adversarial original input data and mean output learning logic determines a mean response for each intermediate layer of the neural network based on results of processing the original input data. The neural network processes adversarial input data and layer-wise comparison logic compares, for each intermediate layer of the neural network, a response generated by the intermediate layer based on processing the adversarial input data, to the mean response associated with the intermediate layer, to thereby generate a distance metric for the intermediate layer. The layer-wise comparison logic generates a vector output based on the distance metrics that is used to train a classifier to identify adversarial input data based on responses generated by intermediate layers of the neural network.Type: GrantFiled: February 1, 2018Date of Patent: March 9, 2021Assignee: International Business Machines CorporationInventors: Gaurav Goswami, Sharathchandra Pankanti, Nalini K. Ratha, Richa Singh, Mayank Vatsa
-
Publication number: 20210049616Abstract: Described are a system, method, and computer program product for real-time payment gateway event monitoring. The method includes receiving input data from a merchant system including an application URL associated with a merchant control interface having a web socket programmed and/or configured to persistently listen for HTTP POST messages, and an event of a payment gateway system. The method also includes monitoring ongoing events of the payment gateway system associated with ongoing transaction messages communicated from point-of-sale (POS) devices to the payment gateway system and detecting the event in the ongoing events. The method further includes, in response to detection of the event, generating a HTTP POST message including event data of the event, and communicating the HTTP POST message to the application URL to cause the merchant control interface to display the event data in a chart for visual representation of the event data.Type: ApplicationFiled: August 14, 2020Publication date: February 18, 2021Inventors: Anurag Gupta, Jagadeesh Saravanan, Richa Singh
-
Patent number: 10901729Abstract: Several API specification files that each include references to code elements that are defined within separate definitions and resources files may be merged together into a single specification for the new API. In this way, specifications for shared code elements that are common across the several specification files may be created without rewriting these specifications each time they are used in an API. A specification merging tool may look for a common reference in the API, match that reference to a definition or resource from the respective definition or resource document, and add that definition or resource to a merged specification file.Type: GrantFiled: March 20, 2019Date of Patent: January 26, 2021Assignee: VISA INTERNATIONAL SERVICE ASSOCIATIONInventors: Richa Singh, Elsi Godolja, Anurag Gupta, Sebastian Amara
-
Publication number: 20200301702Abstract: Several API specification files that each include references to code elements that are defined within separate definitions and resources files may be merged together into a single specification for the new API. In this way, specifications for shared code elements that are common across the several specification files may be created without rewriting these specifications each time they are used in an API. A specification merging tool may look for a common reference in the API, match that reference to a definition or resource from the respective definition or resource document, and add that definition or resource to a merged specification file.Type: ApplicationFiled: March 20, 2019Publication date: September 24, 2020Inventors: Richa Singh, Elsi Godolja, Anurag Gupta, Sebastian Amara
-
Publication number: 20200286093Abstract: A method for providing electronic business updates comprising receiving at least one account update from an account server relating to activity in a user account and determining that the account update is associated with an activity category. Based on the a determination generating an activity notification and, in response to receiving the at least one account update, pushing the activity notification to a user computing device to trigger a notification on the user computing device.Type: ApplicationFiled: March 7, 2019Publication date: September 10, 2020Inventors: Shobhit Agrawal, Anurag Gupta, Richa Singh, Elsi Godolja, Surya Maharjan
-
Publication number: 20190238568Abstract: Mechanisms are provided for training a classifier to identify adversarial input data. A neural network processes original input data representing a plurality of non-adversarial original input data and mean output learning logic determines a mean response for each intermediate layer of the neural network based on results of processing the original input data. The neural network processes adversarial input data and layer-wise comparison logic compares, for each intermediate layer of the neural network, a response generated by the intermediate layer based on processing the adversarial input data, to the mean response associated with the intermediate layer, to thereby generate a distance metric for the intermediate layer. The layer-wise comparison logic generates a vector output based on the distance metrics that is used to train a classifier to identify adversarial input data based on responses generated by intermediate layers of the neural network.Type: ApplicationFiled: February 1, 2018Publication date: August 1, 2019Inventors: Gaurav Goswami, Sharathchandra Pankanti, Nalini K. Ratha, Richa Singh, Mayank Vatsa
-
Publication number: 20190236402Abstract: A neural network is configured to suppress an output of a mitigation node in a mitigation layer of the neural network. The neural network is pre-configured to recognize objects from inputs when operating using a processor and a memory. An actual input is sent to the neural network for object recognition, the actual input is an altered input. By suppressing the output of the mitigation node, the neural network is caused to avoid falsely recognizing an object from the actual input, where the altered input is configured to cause the neural network to falsely recognize the object from the actual input.Type: ApplicationFiled: January 29, 2018Publication date: August 1, 2019Applicants: International Business Machines Corporation, Indraprastha Institute of Information Technology (IIIT), DelhiInventors: Gaurav Goswami, Sharathchandra Pankanti, Nalini K. Ratha, Richa Singh, Mayank Vatsa
-
Patent number: 9145789Abstract: An impingement plate is cooperable with a shroud assembly. The shroud assembly includes an outer shroud and plural inner shrouds with seals between the plural inner shrouds, respectively. The impingement plate includes a trailing edge portion, a leading edge portion and a mid portion between the trailing edge portion and the leading edge portion. A plurality of impingement holes are formed across an area of the impingement plate, and a cooling and damping section includes at least one channel that is shaped to accelerate cooling flow through the impingement plate.Type: GrantFiled: September 5, 2012Date of Patent: September 29, 2015Assignee: GENERAL ELECTRIC COMPANYInventors: Siva Ram Surya Sanyasi Adavikolanu, Ajay Gangadhar Patil, Debdulal Das, Richa Singh
-
Publication number: 20140064913Abstract: An impingement plate is cooperable with a shroud assembly. The shroud assembly includes an outer shroud and plural inner shrouds with seals between the plural inner shrouds, respectively. The impingement plate includes a trailing edge portion, a leading edge portion and a mid portion between the trailing edge portion and the leading edge portion. A plurality of impingement holes are formed across an area of the impingement plate, and a cooling and damping section includes at least one channel that is shaped to accelerate cooling flow through the impingement plate.Type: ApplicationFiled: September 5, 2012Publication date: March 6, 2014Applicant: General Electric CompanyInventors: Siva Ram Surya Sanyasi Adavikolanu, Ajay Gangadhar Patil, Debdulal Das, Richa Singh