Joel Emer

Joel Emer is a Professor of the Practice in EECS, [CS and EE] at the Massachusetts Institute of Technology, affiliated with the Electrical Engineering and Computer Science department. His research interests include Computer Architecture, Integrated Circuits and Systems, and Systems and Networking. Emer's work focuses on advancing technologies in these areas, with recent contributions including developments in secure hardware design and efficient AI task execution. His profile page provides further details on his research and academic activities.

Joel Emer is a Professor of the Practice at MIT and a Sr. Distinguished Research Scientist at Nvidia, specializing in Computer Architecture. His research focuses on optimizing the performance and energy efficiency of computing systems, particularly in the context of deep neural networks. Emer has developed several energy-efficient accelerators for convolutional neural networks, including Eyeriss and SCNN, and has explored novel architectures for mobile and embedded systems. His work also addresses challenges in hardware design for machine learning, such as sparse convolutional neural networks and error propagation in DNN accelerators. Additionally, he has contributed to the understanding of architectural vulnerability factors and cache management techniques.