In the last decade, deep learning has played an indispensable role in the human world. At the core, general matrix multiplication (GEMM) has prompted extensive optimization efforts on conventional binary encoding, to democratize the transformative capabilities. However, binary computing does not offer optimal efficiency. To yield unprecedented hardware efficiency in new applications, my research leverages unconventional computing paradigms to design next-generation computer architecture, including unary, neuromorphic, approximate computing, and beyond. In this talk, I will focus on how unary computing utilizes extremely simple hardware to manipulate unary bitstreams, and how unary computing contributes to improved hardware efficiency at the architecture level for deep learning.

Dr. Di Wu is an assistant professor in the Department of ECE at the University of Central Florida. He earned his PhD from the Department of ECE at the University of Wisconsin-Madison in 2023. His research interests broadly spread out in emerging areas of computer architecture and system, featured in ASPLOS, HPCA, ISCA, MICRO, etc. His research has received many recognitions, including IEEE Micro Top Pick in 2021, ASPLOS Distinguished Artifact Evaluation Award in 2024 and The Harold Peterson Outstanding Dissertation Award in 2024, etc. He served as the program committee member and reviewer of premier conferences and journals on computer architecture, EDA, VLSI, etc., and served on NSF CISE panels. He also organized multiple workshops and contests for emerging computing.