Pierre Paulin
Synopsys, Canada
Challenges and Solutions for Future-Proof Neural Network Accelerators
Abstract
Deep-learning based solutions for embedded vision have emerged as a key application of the growing class of artificial intelligence-based solutions. Convolutional Neural Networks (CNN), have been a key class of deep-learning solution in the field of embedded vision. While most of the early applications of CNN have been on general purpose GPUs or straightforward extensions of existing vector processors, there has been increased interest in more specialized CNN engines in order to provide high-performance with low power, low area and low bandwidth. In addition, due to the rapid rate of innovation in deep learning, this solution must also remain flexible and future-proof. The EV6 family of embedded vision processor and its programmable dedicated CNN engine are used to illustrate the challenges and design trade-offs needed to achieve class leading MACs/cycle at the cost and power of a hardwired solution, while remaining fully programmable. Using automatic mapping tools that compile CNN graphs captured on well-known CNN frameworks like Caffe and Tensorflow, we illustrate tradeoffs in performance, power, bandwidth and accuracy using well-known CNN graphs. We also introduce CNN graph optimization techniques that can be used to further exploit the available computing resources.
Biography
Dr. Pierre G. Paulin is Director of R&D for Embedded Vision at Synopsys. He is responsible for the application development, architecture design and S/W programming tools for embedded vision processors supporting classical computer vision and neural-network based solutions. Prior to this, he was director of System-on-Chip Platform Automation at STMicroelectronics in Canada, working on platform programming tools for multi-processor systems-on-a-chip, targeting computer vision, video codecs and network processors. This followed his previous positions as director of Embedded Systems Technologies for STMicroelectronics in Grenoble, France, and manager of Embedded Software and High-level synthesis tools with Nortel Networks in Canada. His interests include embedded vision, video processing, multi-processor systems, and system-level design. He obtained a Ph.D. from Carleton University, Ottawa, and B.Sc. and M.Sc. degrees from Laval University, Quebec.