The Distinguished Lecture on “Rethinking Data Movement in AI Chips: Fully Analog or Fully Digital?” will take place as follows:
Date: 13 March 2026 (Friday)
Time: 11:30 – 12:30
Venue: Research Building N21, G/F, G013
The speaker is:
Prof. LIN Longyang, Assistant Professor, Southern University of Science and Technology
The Lecture is:
Rethinking Data Movement in AI Chips: Fully Analog or Fully Digital?
Abstract:
The efficiency of modern artificial intelligence systems is increasingly constrained by the energy cost of moving data between sensors, processors, and memory. Addressing this challenge requires rethinking conventional computing architectures and exploring designs that more tightly integrate sensing, memory, and computation.
This seminar presents two contrasting hardware approaches that explore this goal from opposite ends of the design spectrum. The first part introduces a fully analog sensing-to-inference vision system-on-chip, which maintains a continuous analog datapath from image sensing to multi-layer neural network inference, eliminating repeated analog-to-digital conversions and enabling highly energy-efficient edge vision processing. The second part presents a fully digital compute-in-memory architecture based on STT-MRAM technology that performs lossless and fully parallel matrix–vector multiplications within memory arrays while preserving digital precision and scalability.
Together, these designs illustrate complementary directions for reducing data movement in AI chips and highlight emerging opportunities for integrating sensing, memory, and computation in future intelligent systems.
Biography:
Dr. Longyang Lin received the Ph.D. degree in the Electrical and Computer Engineering Department from the National University of Singapore in 2018. From 2018 to 2021, he served as a Research Fellow at the same institution. He is currently an Assistant Professor with the School of Microelectronics, Southern University of Science and Technology (SUSTech), Shenzhen, China.
Dr. Lin has authored or co-authored over 70 publications on journals and conference proceedings. His research interests include ultra-low-power circuit design, end-to-end compute-in-memory systems, cryogenic digital circuit design, and hardware security. He serves as an Associate Editor for the IEEE Transactions on VLSI Systems. He has also served as a Technical Program Committee member for APCCAS 2022 and ICTA since 2022, and as an Organizing Committee member for AICAS 2026. He was the recipient of the Takuo Sugano Award for Outstanding Far-East Paper at ISSCC 2022.
