Introduction to Machine Learning and Inference Slides and Transcript

Abstract- Machine learning, and in particular deep learning, has received a great deal of attention in recent years as it has disrupted many fields of electrical engineering and computer science. The success of deep-learning techniques comes from their ability to solve notoriously difficult classification and regression problems. Much of deep learning’s recent successes can be attributed to a virtuous cycle of advances in computing hardware, the availability of huge amounts of labeled data, and development of deeper models. This talk introduces the broad and dynamic field of deep learning for hardware designers. We begin with a brief history and review key innovations that have led to the powerful deep-learning techniques we see today. We will review the different types of learning widely used today with a focus on neural network models for inference applied across a variety of domains. The primary objective of this talk is to help and motivate chip designers to engage in this exciting opportunity and further push the impact of deep learning via hardware-level innovations.
Bio- Gu-Yeon Wei is Gordon McKay Professor of Electrical Engineering and Computer Science in the Paulson School of Engineering and Applied Sciences (SEAS) at Harvard University and currently serves as Area Chair for Electrical Engineering. He received his BS, MS, and PhD degrees in Electrical Engineering from Stanford University. His research interests span multiple layers of a computing system: mixed-signal integrated circuits, computer architecture, and design tools for efficient hardware. His research efforts focus on identifying synergistic opportunities across these layers to develop energy-efficient solutions for a broad range of systems from flapping-wing microrobots to machine learning hardware for IoT devices to large-scale servers.