2023
ASYNC 2023最佳论文提名
2021
ASYNC 2021最佳论文提名
2021
北京市科学技术进步一等奖
2021
集成电路学院题为“A 28nm Configurable Asynchronous SNN Accelerator with Energy-Efficient Learning”的论文被2021年IEEE异步电路会议
(27th IEEE International Symposium on Asynchronous Circuits and Systems, IEEE ASYNC)录用,该会议平均录用率低于30%,要求6-8页篇幅。
该论文是清华大学首次作为第一作者单位被该会议录用的论文。论文设计了一款低片上学习代价的异步可重构脉冲神经网络(SNN)加速器,
作者提出并实现了一种稀疏目标传输(S-TP)训练算法,在NMNIST数据集上实现了95.7%的识别精度和3.97 pJ/SOP@0.9V的能效,训练能耗只占总能耗的0.63%。
和最新研究成果JSSC2020[1]与ISSCC2019[2]的工作相比,该加速器芯片的能耗降低了7.5倍,额外训练能耗降低了15倍。
参考文献:
[1] J. Park, J. Lee and D. Jeon, "A 65-nm Neuromorphic Image Classification Processor With Energy-Efficient Training Through Direct Spike-Only Feedback," in IEEE Journal of Solid-State Circuits, vol. 55, no. 1, pp. 108-119, Jan. 2020, doi: 10.1109/JSSC.2019.2942367.
[2] J. Park, J. Lee and D. Jeon, "7.6 A 65nm 236.5nJ/Classification Neuromorphic Processor with 7.5% Energy Overhead On-Chip Learning Using Direct Spike-Only Feedback," 2019 IEEE International Solid- State Circuits Conference - (ISSCC), 2019, pp. 140-142, doi: 10.1109/ISSCC.2019.8662398.
2020
Chen Hong and her team won the third prize of
science and technology of Beijing Medical Association in 2020
2019
The second prize in the finals of the
2019 International College Student Brain Computing Contest
2018
2018 International College Student
Brain Computing Competition Final Award of Excellence
2018-2019
Chen Hong team 2018, 2019 international college student brain computing outstanding instructor
2017
Chen Hong, 2017 China Institute of Electronics Science
and Technology Award second prize
2016
The second prize of National Teaching Achievement Award
in 2016
2013
2013 International Congress of Circuits and
Systems ISCAS Best Demonstration Award