Varun Gupta, 44Piscataway, NJ

2011010, May 5, 2011

Oct 29, 2009

12/608028

SIMON BORST - Convent Station NJ, US
Varun Gupta - Pittsburgh PA, US
Anwar I. Walid - Watchung NJ, US

G06F 12/08
G06F 15/173
G06F 12/00

711118, 709223, 711E12001, 711E12017

A content distribution network (CDN) comprising content storage nodes (CSNs) or caches having storage space that preferentially stores more popular content objects.

2023001, Jan 19, 2023

Sep 15, 2022

17/945529

- Atlanta GA, US
Shu Shi - Summit NJ, US
Rittwik Jana - Montville NJ, US
Varun Gupta - Mountain View CA, US

AT&T Intellectual Property I, L.P. - Atlanta GA

H04N 21/4402
H04N 21/434
H04N 21/45
H04N 21/414

Aspects of the subject disclosure may include, for example, receiving, over a communication network, a plurality of requests for frames of video content to provide to a mobile device. Further embodiments can include determining a first portion of the plurality of requests are for pre-fetch frames of the video content, and providing, over the communication network, the pre-fetch frames to the mobile device over a default bearer path. Additional embodiments can include determining a second portion of the plurality of requests are for emergent frames of the video content, and providing, over the communication network, the emergent frames to the mobile device over a dedicated bearer path. Other embodiments are disclosed.

2021034, Nov 4, 2021

May 1, 2020

16/864852

- McLean VA, US
Varun GUPTA - Edison NJ, US
Tyler MAIMAN - Melville NY, US

G06Q 20/20
G06Q 50/00
G06Q 30/02
G06N 20/00

A device may receive, from a point-of-sale (POS) device, transaction data associated with a transaction between a customer and a merchant associated with the POS device. The device may determine a customer email address and other data associated with the transaction, the customer, or the merchant, based on the transaction data and customer data identifying the customer, and may process the customer email address, the other data, and social media data, with a machine learning model, to identify a social media account of the customer. The social media data may include data identifying multiple social media accounts, and the multiple social media accounts may include the social media account of the customer. The device may determine, based on the social media data, whether the social media account of the customer follows the merchant, and may perform actions based on whether the social media account follows the merchant.

2021029, Sep 23, 2021

Mar 8, 2021

17/195272

- McLean VA, US
Brice ELDER - Brooklyn NY, US
Varun GUPTA - Edison NJ, US

Capital One Services, LLC - McLean VA

H04L 29/06
G06N 20/00

Systems and methods for adaptive token verification are disclosed. For example, a system may include at least one memory storing instructions and one or more processors configured to execute the instructions to perform operations. The operations may include training a verification model to verify tokenized requests based on system identifiers. The operations may include receiving a tokenized request from an external system, the request comprising a system identifier of the external system. The operations may include generating output of the verification model based on the system identifier, and, based on the output, performing one of granting the request or blocking the request.

2021026, Aug 26, 2021

May 7, 2021

17/314552

- Atlanta GA, US
Varun Gupta - Morristown NJ, US
Dhruv Gupta - San Ramon CA, US
Rittwik Jana - Montville NJ, US
Jin Wang - Fremont CA, US
Laurie Bigler - Lafayette CA, US
Weihua Ye - Chicago IL, US
Zhengye Liu - Pleasanton CA, US

H04L 12/26
H04L 12/24
G06N 20/00
H04L 5/00

The described technology is generally directed towards reducing latency in a wireless communications network. Radio access network latency data corresponding to a measured latency impact criterion is obtained by a network device of a wireless network. Based on the radio access network latency data, latency guidance data usable by the radio network device to achieve a reduction in communication latency that is experienced by a user equipment is predicted, e.g., by a learned model. The latency guidance data can be used to facilitate a reduction in the communication latency that is experienced by a user equipment.

2021024, Aug 5, 2021

Apr 26, 2021

17/239840

- Atlanta GA, US
John F. Murray - Denville NJ, US
Michael Hwang - New Providence NJ, US
Rittwik Jana - Montville NJ, US
Varun Gupta - Mountain View CA, US

AT&T Intellectual Property I, L.P. - Atlanta GA

H04N 21/238
H04N 21/414
H04N 21/218
H04N 21/222
H04N 21/24

Aspects of the subject disclosure may include, for example, embodiments for determining a first motion-to-update latency of a mobile device in relation to receiving a video content update provided by a video content server. Further embodiments include, responsive to determining that the first motion-to-update latency exceeds a threshold: determining a motion-to-update latency of the mobile device in relation to receiving video content updates from a plurality of edge servers resulting in a plurality of motion-to-update latencies, identifying a second motion-to-update latency from the plurality of motion-to-update latencies that is below the threshold, identifying an edge server associated with the second motion-to-update latency, and transmitting video content to the edge server to mitigate the first motion-to-update latency. The edge server provides a portion of the video content at different time intervals to the mobile device and the video content comprises panoramic video content. Other embodiments are disclosed.

2021022, Jul 22, 2021

Apr 5, 2021

17/221947

- Atlanta GA, US
Shu Shi - Summit NJ, US
Varun Gupta - Morristown NJ, US
John Murray - Denville NJ, US
Rittwik Jana - Montville NJ, US

AT&T Intellectual Property I, L.P. - Atlanta GA

G06K 9/00
G06N 3/08
G06K 9/62
H04N 5/232
H04N 7/18
H04N 19/87
H04N 19/114

A system can obtain a labelled data set, including historic video data and labelled events. The system can divide the labelled data set into historic training/testing data sets. The system can determine, using the historic training data set, a plurality of different parameter configurations to be used by a video encoder to encode a video that includes a plurality of video frames. Each parameter configuration can include a group of pictures (“GOP”) size and a scenecut threshold. The system can calculate an accuracy of event detection (“ACC”) and a filtering rate (“FR”) for each parameter configuration. The system can calculate, for each parameter configuration of the plurality of different parameter configurations, a harmonic mean between the ACC and the FR. The system can then select a best parameter configuration of the plurality of different parameter configurations based upon the parameter configuration that has the highest harmonic mean.

2021006, Mar 4, 2021

Aug 27, 2019

16/552443

- Atlanta GA, US
Shu Shi - Summit NJ, US
Varun Gupta - Morristown NJ, US
John Murray - Denville NJ, US
Rittwik Jana - Montville NJ, US

AT&T Intellectual Property I, L.P. - Atlanta GA

G06K 9/00
G06N 3/08
G06K 9/62
H04N 19/87
H04N 5/232
H04N 7/18

A system can obtain a labelled data set, including historic video data and labelled events. The system can divide the labelled data set into a historic training/testing data sets. The system can determine, using the historic training data set, a plurality of different parameter configurations to be used by a video encoder to encode a video that includes a plurality of video frames. Each parameter configuration can include a group of pictures (“GOP”) size and a scenecut threshold. The system can calculate an accuracy of event detection (“ACC”) and a filtering rate (“FR”) for each parameter configuration. The system can calculate, for each parameter configuration of the plurality of different parameter configurations, a harmonic mean between the ACC and the FR. The system can then select a best parameter configuration of the plurality of different parameter configurations based upon the parameter configuration that has the highest harmonic mean.