Wei-Chen Wang 王韋程
Postdoctoral Research Associate at MIT EECS
I am currently a Postdoctoral Research Associate at HAN Lab of the Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, working with Prof. Song Han.
I received my Ph.D. degree in Computer Science and Information Engineering from National Taiwan University in 2021, advised by Prof. Tei-Wei Kuo and Prof. Yuan-Hao Chang. Previously, I received my B.S. and M.S. degrees in Computer Science and Information Engineering from National Taiwan University in 2015 and 2017, respectively. Before joining MIT in 2022, I served as an Executive Engineer at Emerging System Laboratory, Macronix International Co., Ltd., which is the largest NOR flash memory manufacturer in the world.
My research interests include efficient deep learning, TinyML, embedded systems, operating systems and memory/storage systems. I have published 21 research papers, which were mainly published in top journals (including premier ACM/IEEE Transactions, e.g., IEEE TC, IEEE TCAD, and ACM TECS) and top conferences (including NeurIPS, IEEE ISSCC, ACM/IEEE MICRO, ACM/IEEE ICCAD, ACM/IEEE ISLPED, and ACM/IEEE CODES+ISSS). Besides, I was the recipient of the 2019 ACM/IEEE CODES+ISSS Best Paper Award and the 2021 IEEE NVMSA Best Paper Award. I have also published 30 US, China, and Taiwan patents, 17 of which have been granted. I am a Member of IEEE and a Member of ACM.
News
- [08/2023] We are excited to opensource TinyChatEngine: On-Device Large Language Model Inference Library. Feel free to give it a try!
- [07/2023] Our paper “PokeLLM: Pocket-sized Engine for Efficient Large Language Models Training” has been accepted by ACM/IEEE MICRO 2023!
- [05/2023] Our paper “Tiny Machine Learning: Progress, Trends and Futures” has been accepted by IEEE Circuits and Systems Magazine (CASM).
- [05/2023] Received the Best Poster Award (First Place) of the NSF Athena AI Institute Annual Review Meeting.
- [04/2023] Our paper “Reaping Both Latency and Reliability Benefits with Elaborate Sanitization Design for 3D TLC NAND Flash” has been accepted by IEEE Transactions on Computers (TC).
- [11/2022] Our team MIT-HAN-Lab receives 1st place (in 150 Teams) in the memory occupation track of ACM/IEEE TinyML Design Contest at ICCAD 2022. [Project]
- [10/2022] Our research project “MCUNetV3: On-Device Training Under 256KB Memory” is highlighted on the MIT homepage!
- [09/2022] Our new paper “On-Device Training Under 256KB Memory” will appear at NeurIPS 2022. It enables tiny on-device training for IoT devices. [Project] [Demo]
- [09/2022] Our paper “ICE: An Intelligent Cognition Engine with 3D NAND-based In-Memory Computing for Vector Similarity Search Acceleration” has been accepted to ACM/IEEE MICRO 2022.
- [08/2022] We are excited to opensource TinyEngine: a memory-efficient and high-performance neural network library for microcontrollers.
- [08/2022] We’re opening a new course on TinyML and Efficient Deep Learning: efficientml.ai.
- [02/2022] Our paper “A 512Gb In-Memory-Computing 3D-NAND Flash Supporting Similar-Vector-Matching Operations on Edge-AI Devices” has been accepted to IEEE ISSCC 2022.
- [07/2021] Our paper “Approximate Programming Design for Enhancing Energy, Endurance and Performance of Neural Network Training on NVM-based Systems” won the Best Paper Award of IEEE NVMSA 2021.
- [10/2019] Our paper “Achieving Lossless Accuracy with Lossy Programming for Efficient Neural-Network Training on NVM-Based Systems” won the Best Paper Award of IEEE/ACM CODES+ISSS 2019
Publications
Journal Papers
- Tiny Machine Learning: Progress, Trends and Futures
Ji Lin, Ligeng Zhu, Wei-Ming Chen, Wei-Chen Wang, and Song Han
IEEE Circuits and Systems Magazine (CASM) (To Appear)
[Paper] - Reaping Both Latency and Reliability Benefits with Elaborate Sanitization Design for 3D TLC NAND Flash
Wei-Chen Wang, Chien-Chung Ho, Yung-Chun Li, Liang-Chi Chen, and Yu-Ming Chang
IEEE Transactions on Computers (TC)
[Paper] - Achieving Lossless Accuracy with Lossy Programming for Efficient Neural-Network Training on NVM-Based Systems
Wei-Chen Wang, Yuan-Hao Chang, Tei-Wei Kuo, Chien-Chung Ho, Yu-Ming Chang, and Hung-Sheng Chang
ACM Transactions on Embedded Computing Systems (TECS), vol. 18, no. 5s, pp. 68:1-68:22, Oct. 2019. (Integrated with ACM/IEEE CODES+ISSS’19) (Best Paper Award)
[Paper] [Slides] [Poster]- Media Coverage: NTU Spotlight News
- Scrubbing-aware Secure Deletion for 3D NAND Flash
Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo, and Ping-Hsien Lin
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), vol. 37, no. 11, pp. 2790-2801, Nov. 2018. (Integrated with ACM/IEEE CODES+ISSS’18)
[Paper] [Slides] [Poster]
Conference Papers
- PokeLLM: Pocket-sized Engine for Efficient Large Language Models Training
Ligeng Zhu, Lanxiang Hu, Ji Lin, Wei-Ming Chen, Wei-Chen Wang, Chuang Gan, and Song Han
ACM/IEEE International Symposium on Microarchitecture (MICRO), 2023.
[Paper] [Slides] - Retention Leveling: Leverage Retention Refreshing and Wear Leveling Techniques to Enhance Flash Reliability with the Awareness of Temperature
Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo, and Yu-Ming Chang
IEEE Nonvolatile Memory Systems and Applications Symposium (NVMSA), 2023.
[Paper] [Slides] - Efficient Sanitization Design for LSM-based Key-Value Store over 3D MLC NAND Flash
Liang-Chi Chen, Shu-Qi Yu, Chien-Chung Ho, Wei-Chen Wang, and Yung-Chun Li
ACM/SIGAPP Symposium on Applied Computing (SAC), 2023.
[Paper] - On-Device Training Under 256KB Memory
Ji Lin, Ligeng Zhu, Wei-Ming Chen, Wei-Chen Wang, Chuang Gan, Song Han
Conference on Neural Information Processing Systems (NeurIPS), 2022.
[Paper] [Project Page] [Code] [Demo Video]- Media Coverage: MIT Spotlight News / MarkTechPost / Analytics Insight / Analytics India Magazine / The Register / Semiconductor Engineering / IoT M2M Council (IMC)
- ICE: An Intelligent Cognition Engine with 3D NAND-based In-Memory Computing for Vector Similarity Search Acceleration
Han-Wen Hu, Wei-Chen Wang, Yuan-Hao Chang, Yung-Chun Lee, Bo-Rong Lin, Huai-Mu Wang, Chong-Ying Lee, Yu-Ming Huang, Yen-Po Lin, Tzu-Hsiang Su, Chih-Chang Hsieh, Chia-Ming Hu, Yi-Ting Lai, Chung-Kuang Chen, Han-Sung Chen, Hsiang-Pang Li, Tei-Wei Kuo, Keh-Chung Wang, Meng-Fan Chang, Chun-Hsiung Hung, and Chih-Yuan Lu
ACM/IEEE International Symposium on Microarchitecture (MICRO), 2022.
[Paper] [Slides] - RNA-seq Quantification on Processing in memory Architecture: Observation and Characterization
Liang-Chi Chen, Shu-Qi Yu, Chien-Chung Ho, Yuan-Hao Chang, Da-Wei Chang, Wei-Chen Wang, and Yu-Ming Chang
IEEE Nonvolatile Memory Systems and Applications Symposium (NVMSA), 2022.
[Paper] [Slides] - RAM: Exploiting Restrained and Approximate Management for Enabling Neural Network Training on NVM-based Systems
Chien-Chung Ho, Wei-Chen Wang, Szu-Yu Chen, Yung-Chun Li and Kun-Chi Chiang
ACM/SIGAPP Symposium On Applied Computing (SAC), 2022.
[Paper] [Slides] - On Enduring More Data Through Enabling Page Rewrite Capability on Multi-level-cell Flash Memory
Yu-Ming Chang, Chien-Chung Ho, Che-Wei Tsao, Shu-Hsien Liao, Wei-Chen Wang, Tei-Wei Kuo, and Yuan-Hao Chang
ACM/SIGAPP Symposium on Applied Computing (SAC), 2022.
[Paper] [Slides] - A 512Gb In-Memory-Computing 3D NAND Flash Supporting Similar Vector Matching Operations on AI Edge Devices
Han Wen Hu, Wei-Chen Wang, Chung-Kuang Chen, Yung-Chun Lee, Bo-Rong Lin, Huai-Mu Wang, Yen-Bo Lin, Yu-Chao Lin, Chih-Chang Hsieh, Chia-Ming Hu, Yi-Ting Lai, Yuan-Hao Chang, Hsiang-Pang Li, Han-Sung Chen, Tei-Wei Kuo, Keh-Chung Wang, Meng-Fan Chang, Chun-Hsiung Hung, and Chih-Yuan Lu
IEEE International Solid-State Circuits Conference (ISSCC), 2022.
[Paper] [Slides] [Poster] - Approximate Programming Design for Enhancing Energy, Endurance and Performance of Neural Network Training on NVM-based Systems
Chien-Chung Ho, Wei-Chen Wang, Te-Hao Hsu, Zhi-Duan Jiang, and Yung-Chun Li
IEEE Nonvolatile Memory Systems and Applications Symposium (NVMSA), 2021. (Best Paper Award)
[Paper] [Slides] - Space-efficient Graph Data Placement to Save Energy of ReRAM Crossbar
Ting-Hsuan Lo, Chun-Feng Wu, Yuan-Hao Chang, Tei-Wei Kuo, and Wei-Chen Wang
ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED), 2021.
[Paper] [Slides] - Challenges and Designs for Secure Deletion in Storage Systems
Wei-Chen Wang, Chien-Chung Ho, Yu-Ming Chang, and Yuan-Hao Chang
IEEE International Conference on Computing, Analytics and Networks (ICAN), 2020.
[Paper] [Slides] - Toward Instantaneous Sanitization through Disturbance-induced Errors and Recycling Programming over 3D Flash Memory
Wei-Chen Wang, Ping-Hsien Lin, Yung-Chun Li, Chien-Chung Ho, Yu-Ming Chang, and Yuan-Hao Chang
ACM/IEEE International Conference on Computer-Aided Design (ICCAD), 2019.
[Paper] [Slides] - Achieving Lossless Accuracy with Lossy Programming for Efficient Neural-Network Training on NVM-Based Systems
Wei-Chen Wang, Yuan-Hao Chang, Tei-Wei Kuo, Chien-Chung Ho, Yu-Ming Chang, and Hung-Sheng Chang
ACM/IEEE International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS), 2019. (Journal Track, Integrated with ACM TECS) (Best Paper Award)
[Paper] [Slides] [Poster]- Media Coverage: NTU Spotlight News
- Achieving Fast Sanitization with Zero Live Data Copy for MLC Flash Memory
Ping-Hsien Lin, Yu-Ming Chang, Yung-Chun Li, Wei-Chen Wang, Chien-Chung Ho, and Yuan-Hao Chang
ACM/IEEE International Conference on Computer-Aided Design (ICCAD), 2018.
[Paper] [Slides] - Scrubbing-aware Secure Deletion for 3D NAND Flash
Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo, and Ping-Hsien Lin
ACM/IEEE International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS), 2018. (Journal Track, Integrated with IEEE TCAD)
[Paper] [Slides] [Poster] - A Stride-away Programming Scheme to Resolve Crash Recoverability and Data Readability Issues of Multi-level-cell Flash Memory
Chien-Chung Ho, Yung-Chun Li, Ping-Hsien Lin, Wei-Chen Wang, and Yuan-Hao Chang
IEEE Nonvolatile Memory Systems and Applications Symposium (NVMSA), 2018.
[Paper] [Slides]
Patents and Patent Applications
US Patents
- Memory Device and Operation Method Thereof
Wei-Chen Wang, Ting-Hsuan Lo, Chun-Feng Wu, Yuan-Hao Chang, Tei-Wei Kuo
US 11,640,255 B2; Issued May 2, 2023.
[Patent] - Memory Device and Wear Leveling Method for the Same
Wei-Chen Wang, Hung-Sheng Chang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo
US 11,550,709 B2; Issued Jan 10, 2023.
[Patent] - Memory Device for Neural Networks
Wei-Chen Wang, Hung-Sheng Chang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo
US 11,526,285 B2; Issued Dec 13, 2022.
[Patent] - Data Management Method for Memory and Memory Apparatus Using the Same
Yung-Chun Li, Wei-Chen Wang
US 11,222,693 B2; Issued Jan 11, 2022.
[Patent] - Memory System, Method of Operating Memory, and Non-Transitory Computer Readable Storage Medium
Ping-Hsien Lin, Wei-Chen Wang, Hsiang-Pang Li, Shu-Hsien Liao, Che-Wei Tsao, Yuan-Hao Chang, Tei-Wei Kuo
US 11,194,515 B2; Issued Dec 7, 2021.
[Patent] - Memory Management Apparatus and Memory Management Method
Wei-Chen Wang, Ping-Hsien Lin, Tse-Yuan Wang, Yuan-Hao Chang, Tei-Wei Kuo
US 11,042,308 B2; Issued Jun 22, 2021.
[Patent] - Memory Device and Operating Method Thereof
Wei-Chen Wang, Han-Wen Hu, Yung-Chun Li, Huai-Mu Wang, Chien-Chung Ho, Yuan-Hao Chang, and Tei-Wei Kuo
US 2023/0221956 A1; Published Jul 13, 2023.
[Patent] - Memory Device and Operating Method Thereof
Wei-Chen Wang, Han-Wen Hu, Yung-Chun Li, Huai-Mu Wang, Chien-Chung Ho, Yuan-Hao Chang, and Tei-Wei Kuo
US 2023/0221882 A1; Published Jul 13, 2023.
[Patent] - Ternary Content Addressable Memory (TCAM)-Based Training Method for Graph Neural Network and Memory Device
Using the Same
Wei-Chen Wang, Yu-Bang Wang, Yuan-Hao Chang, and Tei-Wei Kuo
US 2023/0162024 A1; Published May 25, 2023.
[Patent] - Memory Device and Operation Method Thereof
Han-Wen Hu, Yung-Chun Li, Bo-Rong Lin, Huai-Mu Wang, and Wei-Chen Wang
US 2023/0161556 A1; Published May 25, 2023.
[Patent] - In-Memory Computing Method and Apparatus
Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo
US 2021/0326114 A1; Published Oct 21, 2021.
[Patent] - Operation Method for Artificial Neural Network
Wei-Chen Wang, Shu-Yin Ho, Chien-Chung Ho, Yuan-Hao Chang
US 2021/0158160 A1; Published May 27, 2021.
[Patent]
China Patents
- Memory Device and Data Management Method Applied Thereto
Ping-Hsien Lin, Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo, Yu-Ming Chang
CN 109710173 B; Issued Dec 3, 2021.
[Patent] - Training Method of Graphic Neural Network Using Ternary Content Addressing Memory and Memory Device
Using the Same
Wei-Chen Wang, Yu-Bang Wang, Yuan-Hao Chang, and Tei-Wei Kuo
CN 116167405 A; Published May 26, 2023.
[Patent] - Memory Device and Method of Operating the Same
Han-Wen Hu, Yung-Chun Li, Bo-Rong Lin, Huai-Mu Wang, and Wei-Chen Wang
CN 116153367 A; Published May 23, 2023.
[Patent] - Memory Device
Yung-Chun Li, Wei-Chen Wang
CN 113741799 A; Published Dec 3, 2021.
[Patent] - In-Memory Operation Method and Device
Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo
CN 113537453 A; Published Oct 22, 2021.
[Patent] - Operation Method of Neural Network
Wei-Chen Wang, Shu-Yin Ho, Chien-Chung Ho, Yuan-Hao Chang
CN 112948753 A; Published Jun 11, 2021.
[Patent] - Memory System and Memory Operation Method
Ping-Hsien Lin, Wei-Chen Wang, Hsiang-Pang Li, Shu-Hsien Liao, Che-Wei Tsao, Yuan-Hao Chang, Tei-Wei Kuo
CN 112506809 A; Published Mar 16, 2021.
[Patent]
Taiwan Patents
- Memory Device and Operating Method Thereof
Wei-Chen Wang, Han-Wen Hu, Yung-Chun Li, Huai-Mu Wang, Chien-Chung Ho, Yuan-Hao Chang, and Tei-Wei Kuo
TW I806641 B; Issued Jun 21, 2023.
[Patent] - Memory Device and Operating Method Thereof
Wei-Chen Wang, Han-Wen Hu, Yung-Chun Li, Huai-Mu Wang, Chien-Chung Ho, Yuan-Hao Chang, and Tei-Wei Kuo
TW I806640 B; Issued Jun 21, 2023.
[Patent] - Ternary Content Addressable Memory (TCAM)-Based Training Method for Graph Neural Network and Memory Device
Using the Same
Wei-Chen Wang, Yu-Bang Wang, Yuan-Hao Chang, and Tei-Wei Kuo
TW I799171 B; Issued Apr 11, 2023.
[Patent] - Memory Device and Operation Method Thereof
Han-Wen Hu, Yung-Chun Li, Bo-Rong Lin, Huai-Mu Wang, and Wei-Chen Wang
TW I796977 B; Issued Mar 21, 2023.
[Patent] - In-Memory Computing Method and In-Memory Computing Apparatus
Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo
TW I792247 B; Issued Feb 11, 2023.
[Patent] - Memory Device and Memory Operation Method
Wei-Chen Wang, Yung-Chun Li, Shu-Yin Ho, Chien-Chung Ho
TW I780967 B; Issued Oct 11, 2022.
[Patent] - Operation Method for Artificial Neural Network
Wei-Chen Wang, Shu-Yin Ho, Chien-Chung Ho, Yuan-Hao Chang
TW I770668 B; Issued Jul 11, 2022.
[Patent] - Memory Apparatus
Yung-Chun Li, Wei-Chen Wang
TW I747532 B; Issued Nov 21, 2021.
[Patent] - Memory System and Method of Operating Memory
Ping-Hsien Lin, Wei-Chen Wang, Hsiang-Pang Li, Shu-Hsien Liao, Che-Wei Tsao, Yuan-Hao Chang, Tei-Wei Kuo
TW I694449 B; Issued May 21, 2020.
[Patent] - Memory Device and Data Management Method Thereof
Ping-Hsien Lin, Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang, Tei-Wei Kuo, Yu-Ming Chang
TW I629592 B; Issued Jul 11, 2018.
[Patent] - Neural Network System and Memory Operation Method
Wei-Chen Wang, Chien-Chung Ho, Yuan-Hao Chang
TW 202230220 A; Published Aug 1, 2022.
[Patent]
Honors and Awards
- 2023 Best Poster Award (First Place) of the NSF Athena AI Institute Annual Review Meeting
- 2022 First Place (in 150 Teams) in the Memory Occupation Track of ACM/IEEE TinyML Design Contest at ICCAD
- 2022 Third Place (in 150 Teams) in the Overall Track of ACM/IEEE TinyML Design Contest at ICCAD
- 2022 Postdoctoral Research Abroad Fellowship sponsored by Taiwan National Science and Technology Council (NSTC)
- 2021 Outstanding Ph.D. Dissertation Award of Institute of Information and Computing Machinery (IICM)
- 2021 Best Ph.D. Dissertation Award of Taiwan Information Storage Association (TISA)
- 2021 Best Paper Award of IEEE NVMSA
- 2021 Graduate Students Study Abroad Fellowship elected by Taiwan National Science and Technology Council (NSTC)
- 2021 Taiwan Ministry of the Interior Award for Research and Development Alternative Service
- 2020 Student Outstanding Performance Award of National Taiwan University (NTU)
- 2020 Taiwan Semiconductor Industry Association (TSIA) Award
- 2020 Best Master Thesis Award of Taiwan Information Storage Association (TISA)
- 2020 Outstanding Research Performance Scholarship of CTCI Foundation
- 2020 NTUEE-1975 Innovation and Entrepreneurship Fund Award
- 2019 Scholarship of Pan Wen Yuan Foundation
- 2019 Best Paper Award of ACM/IEEE CODES+ISSS
- 2019 Student Travel Grant Award of Embedded Systems Week (ESWEEK)
- 2015 Fourth Prize (in 821 teams) of ACM KDD Cup
Academic Services
TPC/PC Member
- Program Committee Member, ACM/SIGAPP Symposium on Applied Computing (SAC), 2023.
- Program Committee Member, ACM/SIGAPP Symposium on Applied Computing (SAC), 2022.
Conference Reviewer
- Conference on Machine Learning and Systems (MLSys), 2023.
- IEEE International Symposium on High-Performance Computer Architecture (HPCA), 2022.
- ACM/IEEE Design Automation Conference (DAC), 2022, 2021 and 2020.
- ACM/IEEE International Conference on Computer-Aided Design (ICCAD), 2022.
- IEEE Real-Time Systems Symposium (RTSS), 2021 and 2019.
- IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), 2021.
- ACM/IEEE Asia and South Pacific Design Automation Conference (ASP-DAC), 2021 and 2019.
- ACM/IEEE International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS), 2020.
- IEEE International Conference on Multimedia Big Data (BigMM), 2019.
- International Workshop on Next-Generation Operating Systems for Cyber-Physical Systems (NGOSCPS), 2019.
- ACM Research in Adaptive and Convergent Systems (RACS), 2018.
- IEEE Non-Volatile Memory Systems and Applications Symposium (NVMSA), 2018.
Journal Reviewer
- IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
- IEEE Transactions on Emerging Topics in Computing (TETC)
- IEEE Transactions on Computers (TC)
- IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD)
- ACM Transactions on Storage (TOS)
- IEEE Transactions on Very Large Scale Integration Systems (TVLSI)
Experiences
- Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology (MIT)
- Postdoctoral Research Associate (Mar. 2023 – Present)
- Postdoctoral Research Fellow (Mar. 2022 – Feb. 2023)
- Research topics: Enabling efficient large language models, deep learning and TinyML on edge devices.
- TinyChatEngine: On-Device Large Language Model Inference Library
- Tiny Training Engine: Pocket-sized Engine for Efficient On-device Training
- MCUNetV3: On-Device Training Under 256KB Memory
- TinyEngine: A Memory-Efficient and High-Performance Neural Network Library for Microcontrollers
- Emerging System Laboratory, Macronix International Co., Ltd.
- Executive Engineer (Jul. 2021 – Feb. 2022)
- Principal Engineer (Sep. 2019 – Jun. 2021)
- Senior Engineer (Sep. 2017 – Aug. 2019)
- Research topics:
- Memory/storage systems: e.g., controller of NAND flash, and data sanitization
- Emerging nonvolatile memory and storage technologies: e.g., phase-change memory
- Next-generation memory/storage architecture designs: e.g., in-memory computing
- Neural networks: e.g., convolutional neural networks, and recurrent neural networks