High performance power efficient neural network implementations on embedded devices

#Neural #Networks #Embedded #Devices
Share

There’s a general trend towards the increased instrumentation of spaces, objects and people. Locally extracting information from sensor data and acting on it has advantages in terms of latency, energy, reliability and security. xNNs can be practically designed and trained to map from data to information for a variety of common sensor types. This talk will cover methods for high performance power efficient xNN implementations on embedded devices and their application to vision, language, speech and controls problems.


  Date and Time

  Location

  Hosts

  Registration


  • Date: 21 Nov 2019
  • Time: 10:00 PM UTC to 11:00 PM UTC
  • Add_To_Calendar_icon Add Event to Calendar
  • Rice University, ECE Department
  • 6100 Main St.
  • Houston, Texas
  • United States 77005
  • Building: Abercrombie Engineering Laboratory
  • Room Number: A-227 Second Floor Conference Room
  • Contact Event Host

  • Prof. Joseph Cavallaro, Houston CASS Chapter Chair, and Prof. Thanh Tran at Rice U.

  • Co-sponsored by Rice ECE Department

  Speakers

Arthur J. Redfern of Texas Instruments

Topic:

High performance power efficient neural network implementations on embedded devices

There’s a general trend towards the increased instrumentation of spaces, objects and people. Locally extracting information from sensor data and acting on it has advantages in terms of latency, energy, reliability and security. xNNs can be practically designed and trained to map from data to information for a variety of common sensor types. This talk will cover methods for high performance power efficient xNN implementations on embedded devices and their application to vision, language, speech and controls problems.

Biography:

Arthur J. Redfern received a B.S. in 1995 from the University of Virginia and a M.S. and Ph.D. in 1996 and 1999, respectively, from the Georgia Institute of Technology, all in electrical engineering. Following his thesis work on nonlinear systems modeled by the Volterra series, Arthur joined Texas Instruments where he currently manages the Machine Learning Lab. His activities at TI have spanned the areas of machine learning (neural network based applications, software and hardware), high performance computing (software), signal processing for analog systems (ADCs, amplifiers, DACs, design optimization, speakers and touch screens) and physical layer communication system design (DSL, DTV and SerDes). In addition to his work at TI, Arthur also teaches a graduate special topics course on deep learning in the UT Dallas CS department. He has over 25 papers published in refereed conferences and journals and has been granted over 25 US patents.

Email:

Address:Texas Instruments, , Dallas, Texas, United States




Agenda

Technical Talk by Dr. Arthur J. Redfern, Manager, Machine Learning Lab, Texas Instruments