The Next Challenge in Computing - Efficient Machine Learning Architecture by Uri Weiser, Technion
Wednesday, October 6, 2021
1240 Computer Sciences
Abstract: For the last 40 years Process Technology and Computer Architecture have been orchestrating the magnificent growth in computing performance; Process Technology was the main locomotive, while Computer Architecture contributed for only about a 1/3 of the performance outcome. It seems that we have reached a major turning point; Moore’s law is reaching its end and Dennard scaling has already ended, while performance requirements continue to soar for many new exciting applications. The combination of new “killer applications” (Machine Learning) and the trend towards Heterogeneous computing provide a new challenged thrust in computer architecture.
In this talk I will present the transformative change in computing to support the new “killer applications”. This change in computing based on Machine Learning calls for new heterogeneous architectures. The new huge demand in computing capacity calls for efficient architecture to mitigate the power. I will highlight some of our specific research aiming at techniques to improve Machine Learning Hardware efficiency and its implications.