Deming Chen: A3C3 – AI Algorithm & Accelerator Co-design, Co-search, and Co-generation

In a conceptual paper published in 2018, we proposed a new concept of AI Algorithm & Accelerator Co-design/Co-search/Co-generation method. We then realized and demonstrated this new method through a series of papers using FPGAs and GPUs as accelerators. Especially, one of the DNN models coming out of this new method, called SkyNet, won double championships in the competitive DAC System Design Contest in 2019. In this talk, we will introduce the key concepts of the A3C3 method to demonstrate why it has been so successful, especially when the method is applied to edge devices. Meanwhile, we will share new insights for applying the principle of the A3C3 strategy for accelerating large language models, represented by our recent works Medusa and SnapKV.

Deming Chen is the Abel Bliss Professor in the Grainger College of Engineering at the University of Illinois Urbana-Champaign. His research interests include hybrid cloud systems, machine learning and AI, security and confidential computing, reconfigurable and heterogeneous computing, and system-level design methodologies. He has published ~300 research papers, received 10 Best Paper Awards and an ACM/SIGDA TCFPGA Hall-of-Fame Paper Award, and delivered more than 160 invited talks. His work has had a significant impact, with open-source solutions adopted by industry, such as FCUDA, DNNBuilder, CSRNet, SkyNet, ScaleHLS, and Medusa. Notably, Medusa has been integrated into Nvidia's TensorRT-LLM, improving the speed of large language model (LLM) execution by 1.9-3.6x. He is an IEEE Fellow, an ACM Distinguished Speaker, and the former Editor-in-Chief of ACM Transactions on Reconfigurable Technology and Systems (TRETS). Under his leadership, the impact factor of ACM TRETS has increased by 3.8 times. He serves as the Illinois Director of the IBM-Illinois Discovery Accelerator Institute and the Director of the AMD-Xilinx Center of Excellence. Additionally, he has been involved in several startup companies, including AutoESL and Inspirit IoT. He received his Ph.D. in Computer Science from UCLA in 2005.