|
|
|
|
Professor Kurt Keutzer |
My approach to research has been
driven by a couple of adages: "Necessity is the mother of invention"
and "Research has to be good for something before it is good for
everything." Over my career the aim of my research has been to enable
system designers to fully utilize the capabilities that silicon manufacturing has
provided them. After many years of working on the problem of how to design Integrated Circuits (i.e.
Electronic Design Automation), I turned my focus to how to program Integrated Circuits, particularly on-chip multiprocessor
systems. Since joining Berkeley I have been exploring applications of “personal
high-performance computing” enabled by these on-chip multiprocessors systems.
In particular my group has focused on applications in computer-vision, speech
recognition, multimedia analytics, and computational finance. As Deep Neural
Networks have proved successful at providing state-of-the-art solutions to all
of these problems, I am now focusing my entire research efforts on the problems
of providing state-of-the-art solutions to the training and deployment of Deep
Neural Networks.
I took an unusual route to academia: I joined Berkeley's EECS faculty in January 1998 after fifteen years in industry. My last industry position was CTO and Senior Vice-President at Synopsys. To get a better idea as to what I've been up to you can check out my brief professional biography.