Date：2025/10/16 11:00-12:10
Location：R107, CSIE
Speakers：Prof. Christophe Dubach
Host：楊佳玲教授
Abstract:
In this talk, I will introduce Equality Saturation, an optimization technique proposed in 2009 that is rapidly gaining traction in the compiler community. I will present two applications from my group: 1) discovering code idioms that can be accelerated using high-performance libraries or specialized accelerators, and 2) integrating equality saturation into the rapidly growing MLIR compiler infrastructure. I will then discuss the main challenge of equality saturation: the exponential time it takes to saturate. I will conclude the talk with our very early effort to address this scalability issue using reinforcement learning with graph neural networks.
Biography:
Since 2020, Christophe Dubach is an Associate Professor jointly appointed in the Department of Electrical and Computer Engineering (ECE) and the School of Computer Science (CS) at McGill University, and an associate member at Mila. Prior to that, he was a Reader (Associate Professor) at the University of Edinburgh. He received a PhD in Informatics from the University of Edinburgh in 2009 and an MSc degree in Computer Science from EPFL in 2005. In 2010, he spent a year as a visiting researcher at the IBM Watson Research Center (USA) working on the LiquidMetal project. His current research interests include high-level programming models for heterogeneous systems, co-design of both computer architecture and optimising compiler, high-level synthesis, and the application of machine learning to these areas.