Links
May/2/2024
We have been working on next generation neuromorphic computing and our work about ANN-SNN conversion is accepted by ICML 2024.March/1/2024
Zhezhi He will serve as a program committee member for IEEE/ACM International Conference on Computer-Aided Design.Nov./8/2023
Congrat to Chenyu for his paper titled "PIMLC$:$ Logic Compiler for Bit-serial Based PIM" which is accepted by DATE 2024.Oct./15/2023
Our ICCD paper titled "GIM$:$ Versatile GNN Acceleration with Reconfigurable Processing-in-Memory" receives the best paper nomination!Sep./24/2023
Our collabrated work with Dr. Rui Yang's group is accepted by Nature Communication is officially online. This work is about the low-temperature integration of ReRAM and 2D material MoS_2 as 1T4R high density memory cell.Aug./25/2023
Congrat to Chen Nie and Chenyang Lv for their papers about PIM for GNN and GPT for logic synthesis (AI4EDA) which are accepted by ICCD 2023.June/1/2023
ICRG has one paper titled "VSPIM, SRAM Processing-in-Memory DNN Acceleration via Vector-Scalar Operations" accpeted by IEEE Transcations on Computers.March/1/2023
Zhezhi He will serve as a program committee member for IEEE/ACM International Conference on Computer-Aided Design.Oct./1/2022
Zhezhi He will serve as a program committee member for design automation conference.Oct/1/2022
ICRG get one project working on the Neuromorphic Computing is funded by the Ministry of Science and Technology. Dr. Zhezhi He is the Co-PI of this project.March/5/2022
Zhezhi He will serve as a program committee member for 2022 IEEE/ACM International Conference on Computer-Aided Design.Feb./5/2022
Our paper titled "Self-Terminated Write of Multi-Level Cell ReRAM for Efficient Neuromorphic Computing" receives Best Paper Award from DATE 2022.Feb/2/2022
Zhezhi He will serve as a program committee member for 32nd ACM Great Lakes Symposium on VLSI.Oct./1/2021
Zhezhi He will serve as a technical program committee member for design automation conference 2022.Aug./20/2021
First NSFC funding of ICRG to conduct research about cross-layer design and optimization of versatile process-in-memory system.May/24/2021
Zhezhi He will serve as a TPC member for 40th International Conference On Computer Aided Design (ICCAD) 2021.Apr./12/2021
Congratulation to Nie Chen for his first publication as a senior undergraduate student. The paper's title is "Energy-Efficient Hybrid-RAM with Hybrid Bit-Serial Based VMM Support", working on novel in-memory computing micro-architecture design.Mar./1/2021
One paper working on the fast domain adaption is accepted by the CVPR-2020, whose title is "KSM-Fast Multiple Task Adaption via Kernel-wise Soft Mask Learning".Feb./20/2021
In this work, we propose a ReRAM-based accelerator for graph neural network inference with specific optimization on load-balance.Feb./2/2021
This paper propose a framework to support edge-cloud collabrative computing for mobile vision task, which is collabrated between Ruguster Univeristy, SJTU, Google, Nvidia, Microsoft, USTC.Dec./22/2020
Zhezhi He receives the ¥450K from Explore-X fundation to support his research in neuromorphic computing and machine learning for EDA.Nov./14/2020
One paper working on the adversarial weight attack detection and accuracy recovery has been accpetd by Design, Automation, and Test in Europe (DATE) 2021.Oct./20/2020
Zhezhi He will serve as a TPC member for 58th Design Automation Conference (DAC) 2021.Oct./5/2020
Zhezhi He has joined Department of Computer Science and Engineering in Shanghai Jiao Tong Univeristy as a tenure-track assistant professor.July/7/2020
Zhezhi's Ph.D. dissertation titled "Efficient and Secure Deep Learning Inference System, A Software and Hardware Co-design Perspective" is successfully defended.May/4/2020
Zhezhi has been awarded Engineering Graduate Fellowship in recognition of his extraordinary academic achievements.Apr./23/2020
Zhezhi has been awarded University Graduate Fellowship (UGF) due to his excellent academic performance. He is the only UGF award receipent for school of Electrical, Computer, Engery Engineering (ECEE) at Arizona State University (ASU).Feb./27/2020
One paper is focus on defending the bit-flip based adversarial weight attack via weight reconstruction technique. Another one is to leverage the group lasso based non-uniform pruning to construct dynamic neural network, which could trade-off the inference accuracy and latency of running hardware on-the-fly.Feb./23/2020
Two papers focuses on the bit-flip based adversarial weight attack are accepted by the CVPR-2020. One is about the defense such adversarial weight attack using binarization-aware training. Another is using insert bit-flip based paraemter Trojan, which is known as Trojan attack and backdoor attack.
Links