Plume Design, Inc.

Qualcomm Incorporated

Paypal, Inc.

Hong Fan

Kiran Kumar Edara

Kaixiang Hu

William John McFarland

Sandip HomChaudhuri

Ashok Ranganath

Gangadhar Burra

Pradeep Kumar Yenganti

Jian Lu

Deepak Dhamdhere

Sameer Vaidya

Vishal Sood

Sudhindra Murthy

Viral Modi

Pattabiraman Subramanian

Adam R Hotchkiss

Paul White

Balaji Rengarajan

Evan Jeng

Grahame Andrew Jastrebski

Aman Singla

Nitin S Sharma

Srinivasa Tagirisa

Divakar Viswanathan

Vidya Sagar Durga

Sheena Chawla

Ashwin Maruti Hegde

Systems, methods, and computer program products are directed to machine learning techniques that use a separate embedding layer. This can allow for continuous monitoring of a processing system based on events that are continuously generated. Various events may have corresponding feature data associated with at least one action relating to a processing system. Embedding vectors that correspond to the features are retrieved from an embedding layer that is hosted on a separate physical device or a separate computer system from a computer that hosts the machine learning system. The embedding vectors are processed though the machine learning model, which may then make a determination (e.g. whether or not a particular user action should be allowed). Generic embedding vectors additionally enable the use of a single remote embedding layer for multiple different machine learning models, such as event driven data models.

Optimally compressed feature representation deployment for automated refresh in event driven learning paradigms

This application discusses identifying data processing timeouts in live risk analysis systems. A service provider, such as an electronic transaction processor, may provide a production computing environment that includes a risk analysis system having one or more risk models, which may be machine-learning based. These risk models may be utilized in order to determine whether incoming data processing requests are fraudulent. To test these risk models using production data traffic, an audit computing environment made of a set of machines that do not service production computing environment requests, but that utilize databases and data connections as are used by the production systems. The audit computing environment may thus mirror the risk models and functionality of the production computing environment without the drawbacks of a typical fully separate testing environment. Thus, risk model performance and execution times may be monitored to determine whether any models encounter errors with production data traffic.

Identifying data processing timeouts in live risk analysis systems

Systems and methods implemented through a user device, for claiming network devices for a home network in a distributed Wi-Fi network includes discovering a newly powered on access point; claiming the newly powered on access point for the home network; and relaying the claimed newly powered on access point to a cloud controller via a disjoint network from the home network, wherein the home network is restricted by the cloud controller to only operate with claimed access points. The discovering, the claiming, and the relaying can be performed by a mobile application executed on the user device.

Claiming network devices for a home network in a distributed Wi-Fi network

Systems and methods implemented by a cloud controller for steering Wi-Fi client devices in a Wi-Fi network include determining a client needs to be steered from a first access point to a second access point; selecting a client steering approach from a plurality of client steering approaches based on the client; and causing the client to be steered from the first access point to the second access point based on the selected client steering approach. The determining can be based on optimization performed by the cloud controller based on operational parameters associated with the Wi-Fi network.

Controlling clients in distributed Wi-Fi networks

Systems and methods for gathering data by an access point in a Wi-Fi system for optimization include periodically or based on command from a cloud-based system performing one or more of i) obtaining on-channel scanning data while operating on a home channel and ii) switching off the home channel and obtaining off-channel scanning data for one or more off-channels; and providing measurement data based on one or more of the on-channel scanning data and the off-channel scanning data to the cloud-based system for use in the optimization of the Wi-Fi system, wherein the measurement data comprises one or more of raw data and processed data.

Data gathering to enable the optimization of distributed Wi-Fi networks

In a particular aspect, a method includes performing a first communication operation associated with a first frequency band using an antenna of a wireless device. The first communication operation is initiated by first communication circuitry of the wireless device. The first communication circuitry is associated with a first communication protocol. The method further includes, based on a duration of a second communication operation, performing the second communication operation associated with a second frequency band using the antenna of the wireless device. The second communication operation is initiated by second communication circuitry of the wireless device. The second communication circuitry is associated with a second communication protocol that is different than the first communication protocol. The first frequency band at least partially overlaps the second frequency band.

Sharing an antenna for wireless communication

Systems and methods include a cloud controller communicatively coupled to one or more distributed Wi-Fi networks and configured to manage the one or more distributed Wi-Fi networks. The cloud controller includes a network interface communicatively coupled to the distributed Wi-Fi networks; one or more processors communicatively coupled to the network interface; and memory storing instructions that, when executed, cause the one or more processors to: determine a new topology state for a topology of Wi-Fi network from a current topology state based on management of the Wi-Fi network; cause one or more nodes to change to new associated parent nodes based on the new topology state; cause an update to a configuration of the one or more nodes based on the new topology state; and continue to change additional nodes to new associated parent nodes based on the new topology state until the new topology state is obtained.

Systems and methods for changing topology and firmware in distributed wi-fi networks

In a particular aspect, a method includes receiving, at a long-term evolution (LTE) circuitry of wireless device from a wireless local area network (WLAN) circuitry of the wireless device while the LTE circuitry has control of at least one antenna of the wireless device, a request for control of the at least one antenna. Communications by the LTE circuitry using the at least one antenna corresponds to a first frequency band, communications by the WLAN circuitry using the at least one antenna correspond to a second frequency band, and the first frequency band at least partially overlaps the second frequency band. The method further includes sending a response from the LTE circuitry to the WLAN circuitry based on data included in the request.

Sharing an antenna between LTE-U and WLAN

Growing community of inventors

San Jose, CA, United States of America

Hong Fan