摘要
This talk describes advances in domain-specific acceleration, which have shown to have enduring importance for computer system design. Domain-specific customisation is often key to the development of accelerators for speeding up important applications involving artificial intelligence workloads, enabling adoption of AI for the benefit of various application domains. To illustrate accelerator development targeting AI for science, Prof. Luk will describe the collaboration with high-energy physics researchers at CERN in developing low-latency deep learning accelerator architectures. In particular, Prof. Luk will present a Graph Neural Network accelerator capable of achieving sub-microsecond latency to support online event selection in the Level-1 triggers at the CERN Large Hadron Collider experiments. The efforts to automate the development of such accelerators based on meta-programming techniques will also be covered.
講者
Professor Wayne Luk
Professor of Computer Engineering, Department of Computing, Imperial College London, UK
Wayne Luk is Professor of Computer Engineering at Imperial College London, and the head of the Custom Computing Research Group there. He was a Visiting Professor at Stanford University. His research focuses on foundation and applications of hardware acceleration, reconfigurable systems, and design automation. He was among the first to develop parametric neural network descriptions for hardware acceleration. His work has led to a Research Excellence Award from Imperial College London, and to over 15 awards from various international conferences. He co-founded BlueBee Technologies which was acquired by Illumina. He co-founded ACM Transactions on Reconfigurable Technology and Systems (TRETS) and was the first Editor-in-Chief. He is a Fellow of the Royal Academy of Engineering, a Fellow of the IEEE, and a Fellow of the BCS.
