Date: Wednesday, 8 November 2006, 6:30 PM
Location: SAP LABS, Building D, 3410 Hillview Avenue, Palo Alto, CA (Google Maps | Yahoo! Maps | Mapquest)
Cost: Free and open to all who wish to attend, but membership is only $10/year.

Topic

We are at the beginning of the multicore era. Computers will have increasingly many cores (processors), but there is still no good programming framework for these architectures, and thus no simple and unified way for machine learning to take advantage of the potential speedup afforded by multicore. We devise a certain "summation form," which allows conforming algorithms to be easily parallelized on multicore computers. We adapt Google's map-reduce paradigm to demonstrate this parallel speedup technique on a wide variety of machine learning and computer vision algorithms. We show that this programming framework is pragmatic and easy to learn for taking advantage of multicore parallelism. Our experimental results show basically linear speedup with an increasing number of processors.

About the Speaker

Dr. Gary Rost Bradski is a Principle Engineer leading machine learning and computer vision research for Intel Labs. He has a joint appointment as consulting Professor at Stanford University. His basic goal is to enable and accelerate AI internally and externally. Some external tools he started for this are the Probabilistic Network Library (PNL) and the Open Source Computer Vision Library (OpenCV) available open, for free on Source Forge. The vision library is used around the world and has become a notable part of the commercial Intel performance library (IPP) products. A statistical machine learning library (MLL) is also available at the OpenCV download site.

Gary received a B.S. degree from U.C. Berkeley and a Ph.D. in Cognitive and Neural Systems (mathematical modeling of biological perception) in May, 1994 from Boston University Center for Adaptive Systems. Gary has worked in the fields of: proximity security systems; medical electronics; computerized EEG; and as a quantitative analyst at First Union's Derivatives Trading Group. Gary also worked as a consultant to the medical electronics industry (NeuroSoft) and in assembly line automation. He is on the advisory board of some local startups in ID theft prevention.

Presentation [PPT]

Presentation [PDF]

Back to the DM SIG page