林哲

一、个人简介
林哲助理教授的主要研究方向为电子设计自动化(EDA)、可重构计算、异构计算及软硬件协同设计等,累计发表高水平学术论文10余篇。近年来,以第一作者发表集成电路设计及自动化领域顶级CCF-A类期刊文章3篇,领域重要期刊文章1篇,领域顶会文章4篇,获得专利授权1项。2022年以第一作者获得EDA领域旗舰会议DATE最佳论文奖提名,2019年以第一作者获FPGA领域旗舰会议FCCM最佳论文奖提名。参与鹏城实验室国家级科研项目、香港大学教育资助委员会项目、香港科技大学-工业界合作项目等多个集成电路设计相关重大项目。
二、研究领域
1. 电子设计自动化(EDA)
随着摩尔定律的不断演进,数字集成电路的复杂度和设计规模持续攀升,单一芯片上可集成的晶体管数量已超过百亿。大规模数字集成电路设计通常采用电子设计自动化(electronic design automation)软件辅助开发。本人的工作研究如何提升数字电路前端设计的能效及开发效率,首次针对高层次综合(HLS),提出基于人工智能算法的早期功耗预测方案,使HLS的能效优化效率提高了高达190倍,获得EDA领域旗舰会议DATE 2022最佳论文奖提名。
2. 可重构计算
可重构计算是一种针对目标应用场景灵活重构硬件架构的计算模式,其最具代表性的计算芯片为现场可编程逻辑门阵列,即FPGA。FPGA常作为复杂算法的硬件加速器,在人工智能、无人驾驶、数字孪生等新型应用中展现出广阔的应用前景。本人的研究工作包括面向FPGA的软硬件协同设计范式、新一代高能效FPGA架构及加速器设计、基于FPGA的硬件设计智能与自动优化算法等,获得FPGA领域旗舰会议FCCM 2019最佳论文奖提名。
3.异构计算及软硬件协同设计
随着半导体制造技术接近物理极限,由CPU组成的同构计算系统逐渐无法满足大数据应用日益增长的算力需求,异构计算应运而生。异构计算系统将FPGA、ASIC等设备作为硬件加速器,协同CPU完成计算密集型任务,满足不同应用在性能和功耗等方面的要求。本人的研究工作涵盖CPU-FPGA异构计算系统体系结构、面向新一代大数据应用的软硬件协同算法、后摩尔时代的编程模型等方面。
三、教育背景
2014.09-2020.01,香港科技大学,博士;
2010.09-2014.06,东南大学,学士。
四、工作经历
2023.03至今,中山大学,助理教授、硕士生导师;
2020.03-2023.02,鹏城实验室,助理研究员(2020-2022)、副研究员(2023)。
五、部分代表性成果
[TCAD23] Zhe Lin, Tingyuan Liang, Jieru Zhao, Sharad Sinha and Wei Zhang. “HL-Pow: Learning-Assisted Pre-RTL Power Modeling and Optimization for FPGA HLS.” In IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), 2023.
[TCAD21] Zhe Lin, Sharad Sinha and Wei Zhang. “Hard-ODT: Hardware-Friendly Online Decision Tree Learning Algorithm and System.” In IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), vol. 40, no. 11, pp. 2279-2292, 2021.
[TCAD19] Zhe Lin, Sharad Sinha and Wei Zhang. “An Ensemble Learning Approach for In-Situ Monitoring of FPGA Dynamic Power.” In IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), vol. 38, no. 9, pp. 1661-1674, 2019.
[TMSCS18] Zhe Lin, Sharad Sinha, Hao Liang, Liang Feng and Wei Zhang. “Scalable Light-Weight Integration of FPGA Based Accelerators with Chip Multi-Processors.” In IEEE Transactions on Multi-Scale Computing Systems (TMSCS), vol. 4, no. 2, pp. 152-162, 2018.
[DATE22]Zhe Lin, Zike Yuan, Jieru Zhao, Wei Zhang, Hui Wang and Yonghong Tian. “PowerGear: Early-stage Power Estimation in FPGA HLS via Heterogeneous Edge-Centric GNNs.” In Proc. of IEEE Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 1341-1346, 2022. IEEE DATE 2022最佳论文提名
[FCCM19]Zhe Lin, Sharad Sinha and Wei Zhang. “Towards Efficient and Scalable Acceleration of Online Decision Tree Learning on FPGA.” In Proc. of IEEE Field-Programmable Custom Computing Machines (FCCM), pp. 172-180, 2019. IEEE FCCM 2019最佳论文提名
[ASPDAC20] Zhe Lin, Jieru Zhao, Sharad Sinha and Wei Zhang. “HL-Pow: A Learning-Based Power Modeling Framework for High-Level Synthesis.” In Proc. of IEEE Asia and South Pacific Design Automation Conference (ASP-DAC), pp. 574-580, 2020.
[FPL17] Zhe Lin, Wei Zhang and Sharad Sinha. “Decision Tree Based Hardware Power Monitoring for Run Time Dynamic Power Management in FPGA.” In Proc. of IEEE International Conference on Field Programmable Logic and Applications (FPL), pp. 1-8, 2017.
[DAC23] Guan Shen, Jieru Zhao, Zeke Wang, Zhe Lin, Wenchao Ding, Chentao Wu, Quan Chen and Minyi Guo. “MARS: Exploiting Multi-Level Parallelism for DNN Workloads on Adaptive Multi-Accelerator Systems.” In Proc. of Design Automation Conference (DAC), to appear, 2023.
[KDD21] Xunqiang Jiang, Tianrui Jia, Yuan Fang, Chuan Shi, Zhe Lin and Hui Wang. “Pre-training on Large-Scale Heterogeneous Graph.” In Proc. of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (KDD), 2021.