A collaborative project with researchers at Oregon State University and University of Southern California received Best Paper Runner-Up Award at a top conference for computer architecture. The research examines if machine learning can also teach us anything about computer architectural design.

Ting-Ru Lin (University of Southern California), Drew Penney (Oregon State University), Massoud Pedram (University of Southern California), Lizhong Chen (Oregon State University) received the Best Paper Runner-Up Award at the International Symposium on High-Performance Computer Architecture on February 26, 2020.

Drew Penney is a doctoral student of electrical and computer engineering, and Lizhong Chen is an assistant professor of electrical and computer engineering at Oregon State.

The paper, “A Deep Reinforcement Learning Framework for Architectural Exploration: A Routerless NoC Case Study,” develops a deep reinforcement learning based framework for flexible and efficient architectural design space exploration. The work demonstrates the viability of utilizing machine learning to improve computer architecture, and the framework will be useful for many researchers in the community.