Embedded Files
Seminars > Seminar Details
by David Z. Pan
Silicon Labs Endowed Chair Professor
Department of Electrical and Computer Engineering
The University of Texas at Austin
Jiaqi Gu
Ph.D. candidate, The University of Texas at Austin
Talk Slides:
In the post-Moore era, conventional electronic digital computing platforms have encountered escalating challenges in supporting massively parallel and energy-hungry artificial intelligence (AI) workloads. Optical neural network (ONN) is a promising hardware platform that could represent a paradigm shift in efficient AI with its ultra-fast speed, high parallelism, and low energy consumption. This talk will give an overview and present cross-layer co-design methodologies for scalable, robust, and self-learnable photonic neural accelerator designs across the circuit, architecture, and algorithm levels. We will introduce area-efficient ONN architecture designs beyond general photonic tensor units with 3-30x lower area cost, efficient ONN on-chip training frameworks that enable in-situ gradient computation with 1,000x higher scalability, and an automated photonic circuit design methodology that surpasses manual designs by 2-30x higher compactness with comparable expressivity and noise robustness.
Speaker Bio:
Prof. David Z. Pan is a Professor and Silicon Laboratories Endowed Chair at the Department of Electrical and Computer Engineering, University of Texas at Austin. His research interests include bidirectional AI and IC interactions, electronic design automation, design for manufacturing, hardware security, and CAD for analog/mixed-signal ICs and emerging technologies. He has published over 450 refereed papers and 8 US patents. He has served in many journal editorial boards and conference committees, including various leadership roles such as DAC 2022 Panel Chair, ICCAD 2019 General Chair, ASP-DAC 2017 TPC Chair, and ISPD 2008 General Chair. He has received many awards, including SRC Technical Excellence Award, 20 Best Paper Awards (from TCAD, DAC, ICCAD, DATE, ASP-DAC, ISPD, HOST, etc.), DAC Top 10 Author Award in Fifth Decade, ASP-DAC Frequently Cited Author Award, ACM/SIGDA Outstanding New Faculty Award, NSF CAREER Award, IBM Faculty Award (4 times), and many international CAD contest awards. He has graduated 43 PhD students and postdocs who have won many awards, including ACM Student Research Competition Grand Finals 1st Place (twice, in 2018 and 2021), and 5 Outstanding PhD Dissertation Awards from SIGDA and EDAA. He is a Fellow of ACM, IEEE and SPIE.
Jiaqi Gu received the B.E. degree in Microelectronic Science and Engineering from Fudan University, Shanghai, China in 2018. He is currently a PhD candidate at the Department of Electrical and Computer Engineering, University of Texas at Austin. His research interests focus on efficient machine learning with emerging technology, next-generation AI with photonic computing, and circuit-architecture-algorithm cross-layer co-design. He has received Best Paper Awards from IEEE TCAD 2021 and ASP-DAC 2020, Best Paper Finalist at DAC 2020, Best Poster Award at NSF Workshop on Machine Learning Hardware (2020), ACM/SIGDA Student Research Competition Gold Medal (2020), and ACM Student Research Competition Grand Finals First Place (2021).
References
Jiaqi Gu, Zheng Zhao, Chenghao Feng, Zhoufeng Ying, Mingjie Liu, Ray T. Chen and David Z. Pan, “Towards Hardware-Efficient Optical Neural Networks: Beyond FFT Architecture via Joint Learnability,” IEEE TCAD 2020.
Jiaqi Gu, Chenghao Feng, Hanqing Zhu, Zheng Zhao, Zhoufeng Ying, Mingjie Liu, Ray T. Chen and David Z. Pan, “SqueezeLight: A Multi-Operand Ring-Based Optical Neural Network with Cross-Layer Scalability,” IEEE TCAD 2022.
Jiaqi Gu, Hanqing Zhu, Chenghao Feng, Zixuan Jiang, Ray T. Chen, and David Z. Pan, “L2ight: Enabling On-Chip Learning for Optical Neural Networks via Efficient in-situ Subspace Optimization,” NeurIPS 2021.
Jiaqi Gu, Hanqing Zhu, Chenghao Feng, Zixuan Jiang, Mingjie Liu, Shuhan Zhang, Ray T. Chen, and David Z. Pan, “ADEPT: Automatic Differentiable DEsign of Photonic Tensor Cores,” ACM/IEEE DAC 2022.
Google Sites
Report abuse