Prof. Shouyi Yin

IEEE Fellow, Vice Director of School of Integrated Circuits
Tsinghua University, China

Speech Title: Reconfigurable Processing Unit: An Unified Architecture for Singal Processing and AI Computing

Abstract: A Reconfigurable processing units originated in signal processing and telecommunications, where their adaptability proved valuable. Today, these architectures provide a unified solution that can efficiently handle both traditional signal processing tasks and modern AI workloads, while achieving improved processing speeds and reduced power consumption. The unified architecture typically incorporates multiple reconfiguration hierarchies: chip-level, processing element array-level, and processing element-level reconfigurations. Chip-level reconfiguration dynamically adjusts the parallelism of multi-chip systems to minimize computation latency and data access. Processing element array-level reconfiguration changes the dataflow or mapping of the computing engine to fully reuse the on-chip data, reducing the memory access. Processing element-level reconfiguration changes the function of the computing unit, such as computing precision and sparsity processing pattern, to increase the bit-wise hardware utilization. This tutorial traces the evolution from signal processing to AI computing, explores the fundamental concepts of reconfigurable technology, discusses its applications in both digital and analog ML processors, and prospects for future development trends in reconfigurable technology.

Bio: Shouyi Yin received the B.S., M.S., and Ph.D. degrees in electronic engineering from Tsinghua University, Beijing, China, in 2000, 2002, and 2005, respectively. He has worked with Imperial College, London, U.K., as a Research Associate. He is currently a full professor and the vice director of School of Integrated Circuits in Tsinghua University. His research interests include reconfigurable computing, AI processors and wafer-scale chips. He has published more than 100 journal papers and more than 50 conference papers. He has served as technical program committee member in the top VLSI and EDA conferences such as ISCA, MICRO, HPCA, DAC, ICCAD, ASPDAC, FPGA and A-SSCC. He is the associate editor of ACM TRETS and Integration, the VLSI journal. He is a fellow of IEEE.

 

Prof. Jun Zhou

National High-Level Scholar, National Young Scholar, Director of the Embedded Artificial Intelligence Special Committee of Sichuan Electronics Society
Executive Deputy Director of Sichuan Engineering Technology Research Center for Intelligent Circuits and Systems of Unmanned Systems
University of Electronic Science and Technology of China, China

 

Speech Title: Ultra-Low Power Domain-Specific AI Processor Design for Edge AI Applications

Abstract: Edge AI applications require to embed AI processors in intelligent devices with ultra-low power and miniaturization requirements. However, the power consumption of existing edge AI processors is still high, making it difficult to meet the demand for ultra-low power AI computing in next generation intelligent devices. By designing domain-specific edge AI processors with algorithms and hardware co-optimization, it is possible to significantly reduce AI computing power consumption while meeting the requirements of high accuracy, low cost with certain application flexibility. This talk discusses the design techniques of ultra-low power domain-specific AI processors for edge AI application with several application examples.

Bio: Zhou Jun is a Professor at the University of Electronic Science and Technology of China. He main research topic is ultra-low-power AI processor design for edge AI applications. He has published 100+ papers in top conferences and journals including ISSCC, JSSC, VLSI, HPCA, DAC and CICC. He serves as TPC member of the Digital Architecture & System (DAS) subcommittee of ISSCC, Chair of the Digital Circuits & Systems (DCS) subcommittee of A-SSCC, and Associate Editor of TBioCAS and TVLSI. He also serves as Director of the Embedded Artificial Intelligence Special Committee of Sichuan Electronics Society, and Executive Deputy Director of Sichuan Engineering Technology Research Center for Intelligent Circuits and Systems of Unmanned Systems. He has been awarded the titles of National High-Level Scholar and National Young Scholar (2017).

 

Prof. Weiqiang Liu

IET Fellow, Head of College of Integrated Circuits
Nanjing University of Aeronautics and Astronautics, China

Speech Title: Approximate Computing: from Circuits to Emerging Applications

Abstract: Computing systems are conventionally designed to operate as accurately as possible. However, this trend faces severe technology challenges, such as power dissipation, circuit reliability, and performance. There are a number of pervasive computing applications (such as machine learning, pattern recognition, digital signal processing, communication, robotics, and multimedia), which are inherently error-tolerant or error-resilient, i.e., in general, they require acceptable results rather than fully exact results. Approximate computing has been proposed for highly energy-efficient systems targeting the above-mentioned emerging error-tolerant applications; approximate computing consists of approximately (inexactly) processing data to save power and achieve high performance, while results remain at an acceptable level for subsequent use. This talk starts with the motivation of approximate computing and then it reviews current techniques for approximate hardware designs. This talk will cover the following topics: 1) approximate arithmetic circuit designs; 2) algorithmic and approximate circuit co-design methods; 3) Applications using approximate computing, such as deep neural networks (DNNs) and security are presented in details. Directions for future works in approximate computing will also be provided.

Bio: Weiqiang Liu is currently a Professor and the Head of College of Integrated Circuits, NUAA. His research interest focuses on energy efficient and secure computing integrated circuits and systems. He has published 3 research books and over over 100 IEEE and ACM journals. He has been awarded the prestigious NSFC Distinguished Young Scholar in 2024, NSFC Excellent Young Scholar in 2020, and the Young Scientist Award by Fok Ying Tung Education Foundation in 2022. He is the Vice President of the IEEE Nanotechnology Council (2024-2025) and the Steering Committee Chair of IEEE TVLSI (2023-2024). He serves as the Associate Editors for IEEE TC, TCAS-I, and TETC, the Guest Editor of Proceedings of the IEEE. He is the general chair of IEEE ARITH 2027 and AsianHOST 2025, the program chair of IEEE NANO 2026 and AsianHOST 2023, program co-chair of IEEE NANO 2025, ACM NANOARCH 2022, IEEE ARITH 2020, and the Track Chair of ASP DAC 2025-2026 and GLSVLSI 2022-2024. He is a Fellow of IET and Senor Member of IEEE.

 

Prof. Hongbin Sun

Vice Dean of State Key Laboratory of Human-Machine Hybrid Augmented Intelligence

Xi'an Jiaotong University, China

Speech Title: Embodied Intelligence in Autonomous Driving: Progress, Challenges, and Future Directions

Abstract: This speech explores embodied intelligence in autonomous driving, spanning algorithmic advances, computing architectures, and domain specific chips. Currently, embodied intelligence in autonomous driving has witnessed remarkable progress. In perception, advanced sensors like LiDAR and high-resolution cameras, combined with multi-sensor fusion techniques, enable vehicles to accurately detect surrounding objects and road conditions even in complex scenarios. For and decision-making, the advent of large-scale pre-trained models, especially Vision-Language-Action (VLA) models, provides more intelligent and interpretable decision-making capabilities, enhancing the vehicle's ability to handle various traffic situations. Moreover, reinforcement learning algorithms are being refined to optimize driving strategies. However, challenges remain, such as the complexity of real-world scenarios that are difficult to fully simulate, and the need for more efficient computing architectures. Future research should focus on improving the generalization ability of models, and integrating advanced computing technologies to further promote the development of embodied intelligence in autonomous driving.

Bio: Hongbin Sun received B.S. and Ph.D. degree in electrical engineering / from Xi’an Jiaotong University, Xi’an, China in 2003 and 2009, respectively. Currently he is a Professor of the Institute of Artificial Intelligence and Robotics, Vice Dean of State Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Xi’an Jiaotong University. He has received the Second Prize of National Technological Invention (2007), the First Prize of Shaanxi Provincial Natural Science Award (2024), the First Prize of Natural Science Award of Chinese Association of Automation (2024), and the Qiushi Artificial Intelligence Science and Education Award (2024).