SHELBY THOMAS

LinkedIn | Facebook | CortexSuite

Short Biography

I am a second year MS/PhD student at the University of California, San Diego (UCSD) working in the computer architecture group with Michael B. Taylor. My work involves creating next generation architectures for brain-like algorithms. Some of my work involves identifying patterns in hardware that can be leveraged to optimize a system running computationally intense workloads.

Contact

Office

University of California, San Diego
Computer Science Engineering 3BU3B 3250
CSE Building 3th Floor

Contact Information

sht005@eng.ucsd.edu

Publications

In my first year at UCSD I had the opportunity to lead a small team of graduates to develop an open source benchmark suite for easy to use machine learning and computer vision algorithms. The result was CortexSuite.

  1. Shelby Thomas, Chetan Gohkale, Enrico Tanuwidjaja, Tony Chong, David Lau, Saturnino Garcia, and Michael Bedford Taylor, "CortexSuite: A synthetic brain benchmark suite" IEEE International Symposium for Workload Characterization (IISWC), 2014.

    Architecture Machine Learning Benchmarks

Teaching

Following are my experiences as a teaching assistant or mentor

  • CSE 198 - Research Mentor, UCSD | (January 2015 - Current)
  • CSE240A - Graduate Architecture, UCSD | (September 2014 - December 2014)
  • CSE141 - Undergraduate Architecture, UCSD | (September 2013 - March 2014)
  • Work Experience

    Software Engineering Intern, GOOGLE (June 2015 - September 2015)

  • Developed production level Neural Network algorithms for two large scale services
  • Lead Undergraduate Researcher, ORACLE (December 2012 - June 2013)

  • Led a three man student team to determine the system level bottlenecks of machine learning algorithms (SVM,ECLAT,HOP) and to implement optimizations
  • Obtained up to a 66% performance boost on algorithms using novel techniques
  • Software Engineering Intern, QUALCOMM (June 2012 - September 2012)

  • Created a test automation suite with python and perl for the Android OS which included UI testing, image compare, and post processing
  • Won awards for the creation and implementation of the suite which significantly reduced turnout time for testing teams

  • Software Engineering Intern, ALGOTOCHIP (July 2011 - September 2011)

  • Leveraged java and the spring framework in order to provide a rich front end user experience
  • Automated database creation, backup, recovery, and processing with perl
  • Talks and Awards

  • Invited Talk at IBM Thomas J. Watson Research Center | New York, USA, 2016

  • Poster Presentation at the Center for Future Architecture Research | Michigan, USA, 2014/2015

  • Invited to Amazon PhD Research Symposium | Seattle, USA, 2014

  • UCSD Computer Science and Engineering Masters Research Initiation Award | UCSD, 2014

  • Skills

    Languages and Libraries

  • Proficient - C/C++, Python, Java, Torch
  • Experienced - Matlab, Vivado, Android, Perl, Javascript
  • Enter Review: