Slides available here!

Speaker:

Ren Wu, Baidu Research Lab, USA

Title:

Deep Learning and Heterogeneous Computing

Abstract:

The rise of the internet, especially mobile internet, has accelerated the data explosion - a driving force for the great success of deep learning in recent years. Behind the scenes, the heterogeneous high-performance computing is another key enabler of that success. For example, the custom-built supercomputer, Minwa, dedicated for deep learning training is one of the key enabler for our recent breakthrough on image recognition. Furthermore, to make the intelligence ubiquitous, heterogeneous computing will play even more important roles.

In this talk, we will share some of work we did at Baidu. We will highlight how big data, deep learning, and high-performance/heterogeneous computing can work together with great success.

Bio:

Dr. Ren Wu is a distinguished scientist at Baidu Research. He is leading the effort to push the frontier of deep learning and artificial intelligence (AI) via high-performance heterogeneous computing. His latest work, Deep Image, powered by custom-designed supercomputer dedicate for deep learning, have achieved state-of-the-art performance on image recognition tasks. His dream is to make AI to be omnipotent and omnipresent.

Prior to joining Baidu, Ren served as chief software architect of Heterogeneous System Architecture (HSA) at AMD. Earlier, he was the principal investigator of CUDA Research Center at HP Labs, where he is widely known for his pioneering work in using GPUs to accelerate big data analytics and large-scale machine learning algorithms.

Ren is also known for his early work on artificial intelligence. His Xiangqi (Chinese chess) program was twice world champion and have dominated computer Xiangqi field for more than a decade. He was the first person to perform systematic research computationally on Xiangqi endgames with astonishing discoveries.



* If you wish to modify any information or update your photo, please contact the web chairmpsoc2014@imag.fr