Kang Wei (韦 康)

Post-doctoral Fellow

Department of Computing (COMP)
Faculty of Engineering
The Hong Kong Polytechnic University

Office:PQ 515, PQ Wing, PolyU, Kowloon, Hong Kong SAR.
Email: kangwei@polyu.edu.hk

Short Bio

He is currently a postdoctoral fellow at The Hong Kong Polytechnic University, supervised by Prof. Song Guo, Fellow of IEEE. He received his Ph.D. degree from Nanjing University of Science and Technology, supervised by Prof. Jun Li. Before that, he received the B.S. degree in information engineering from Xidian University, Xian, China, in 2014. From Jun. 2021 to Dec. 2021, he was a research intern (Rhinoceros Bird Project) at Tencent, Shenzhen, supervised by Dr. Yong Cheng.

He mainly focuses on privacy protection and optimization techniques for edge intelligence, such as federated learning, differential privacy and network resource allocation. He has won the 2022 IEEE Signal Processing Society Best Paper Award (IEEE信号处理学会最佳论文), 2022 Wiley China Open Science Author of the Year (Wiley中国开放科学2022年度作者) and the first place of iDASH Privacy and Security Workshop 2021-Secure Genome Analysis Competition.

For more information, please see my [Curriculum Vitae] and [Google Scholar].

Experiences

• Post-doctoral Fellow:  The Hong Kong Polytechnic University,  Jan. 2023 - Now
  Supervisor: Prof. Song Guo, IEEE Fellow.
  

• Research Intern:  Tencent,  Jun. 2021 - Dec. 2021
  Supervisor: Dr. Yong Cheng.
  

• Ph.D.:  Nanjing University of Science and Technology,  Sept. 2017 - Dec. 2022
  Supervisor: Prof. Jun Li.
  

• Engineer:  Panda Group,  Aug. 2014 - Jun. 2016
  

• B.E.:  Xidian University,  Aug. 2010 - Jul. 2014
  

Publications (Selected)

Conferences

• [ICML]  Improving the Model Consistency of Decentralized Federated Learning,
  Yifan Shi, Li Shen, Kang Wei, Yan Sun, Bo Yuan, Xueqian Wang, Dacheng Tao.
  In Proc. The Fortieth International Conference on Machine Learning (ICML), 2023.
  (CCF-A, Top AI Conference)
  

• [CVPR]  Make Landscape Flatter in Differentially Private Federated Learning,
  Yifan Shi, Yingqi Liu, Kang Wei, Li Shen, Xueqian Wang and Dacheng Tao.
  In Proc. IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR), 2023.
  (CCF-A, Top Computer Vision Conference)
  

• [ICPADS]  CluFL: Cluster-driven weighted FL model aggregation strategy,
  Hanchi Shen, Jun Li, Kang Wei, Pengcheng Xia, Sirui Tian, Ming Ding and Zengxiang Li.
  In Proc. IEEE International Conference on Parallel and Distributed Systems, 2022.
  (CCF-C)
  

• [ITW]  Read-voltage Optimization for Finite Code Length in MLC NAND Flash Memory,
  Kang Wei, Jun Li, Lingjun Kong, Feng Shu, Yonghui Li.
  In Proc. IEEE Information Theory Workshop, 2018, pp. 1-5.
  (EI)
  

Journals

• [TDSC]  Covert Model Poisoning Against Federated Learning: Algorithm Design and Optimization,
  Kang Wei, Jun Li, Ming Ding, Chuan Ma, Yo-Seb Jeon, H. Vincent Poor.
  Accepted by IEEE Transactions on Dependable and Secure Computing, 2023.
  (中科院2区, CCF-A)
  

• [TIFS]  Personalized Federated Learning with Differential Privacy and Convergence Guarantee,
  Kang Wei, Jun Li, Ming Ding, Chuan Ma, Wen Chen, Jun Wu, Meixia Tao, H. Vincent Poor.
  Accepted by IEEE Transactions on Information Forensics & Security, 2023.
  (中科院1区, CCF-A)
  

• [TMC]  User-level Privacy-preserving Federated Learning: Analysis and Performance Optimization,
  Kang Wei, Jun Li, Ming Ding, Chuan Ma, Hang Su, Bo Zhang and H Vincent Poor.
  IEEE Transactions on Mobile Computing, vol. 21, no. 9, pp. 3388-3401, 1 Sept. 2022.
  (Excellent academic paper of Chinese Institute of Electronics, ESI Highly Cited Paper, 中科院2区, CCF-A)
  

• [JSAC]  Low-Latency Federated Learning Over Wireless Channels With Differential Privacy,
  Kang Wei, Jun Li, Chuan Ma, Ming Ding, Cailian Chen, Shi Jin, Zhu Han and H. Vincent Poor.
  IEEE Journal on Selected Areas in Communications, vol. 40, no. 1, pp. 290-307, Jan. 2022.
  (中科院1区, CCF-A)
  

• [TIFS]  Federated Learning with Differential Privacy: Algorithms and Performance Analysis,
  Kang Wei, Jun Li, Ming Ding, Chuan Ma, Howard H. Yang, Farhad Farokhi, Shi Jin, Tony Q. S. Quek and H. Vincent Poor.
  IEEE Transactions on Information Forensics and Security, vol. 15, pp. 3454-3469, 2020.
  (IEEE Signal Processing Society Best Paper Award, ESI Highly Cited Paper, 中科院1区, CCF-A)
  

• [TCA&S-II]  Page-based Dynamic Partitioning Scheduling for LDPC Decoding in MLC NAND Flash Memory,
  Kang Wei, Jun Li, Lingjun Kong, Feng Shu, Francis C. M. Lau.
  IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 66, no. 12, pp. 2082-2086, 2019.
  (中科院2区)
  

• [SCIS]  DP-GenFL: A Local Differentially Private Federated Learning System through Generative Data,
  Jun Li, Kang Wei, Chuan Ma and Feng Shu.
  Accepted by SCIENCE CHINA Information Sciences, 2022.
  (Supervisor is the first author, 中科院2区, CCF-A)
  

• [TIFS]  Amplitude-Varying Perturbation for Balancing Privacy and Utility in Federated Learning,
  Xin Yuan, Wei Ni, Ming Ding, Kang Wei (Corresponding Author), Jun Li and H. Vincent Poor.
  Accepted by IEEE Transactions on Information Forensics & Security, 2023.
  (中科院1区, CCF-A)
  

• [TDSC]  RDP-GAN: A R{\'e}nyi-Differential Privacy based Generative Adversarial Network,
  Chuan Ma, Jun Li, Ming Ding, Bo Liu, Kang Wei (Corresponding Author), Jian Weng and H. Vincent Poor.
  Accepted by IEEE Transactions on Dependable and Secure Computing, 2023.
  (中科院2区, CCF-A)
  

• [IET Communications]  DNN-aided Read-voltage Threshold Optimization for MLC Flash Memory with Finite Block Length,
  Cheng Wang, Kang Wei, Lingjun Kong, Long Shi, Zhen Mei, Jun Li and Kui Cai.
  IET Communications, vol. 16, no. 2, pp. 120-130, 2022.
  (2022 Wiley China Open Science Author of the Year, Wiley, SCI)
  

• [TPDS]  Blockchain Assisted Decentralized Federated Learning (BLADE-FL): Performance Analysis and Resource Allocation,
  Jun Li, Yumeng Shao, Kang Wei, Ming Ding, Chuan Ma, Long Shi, Zhu Han and H Vincent Poor.
  IEEE Transactions on Parallel and Distributed Systems, vol. 33, no. 10, pp. 2401-2415, 1 Oct. 2022.
  (中科院2区, CCF-A)
  

• [JSAC]  Low-latency Federated Learning with DNN Partition in Distributed Industrial IoT Networks,
  Xiumei Deng, Jun Li, Chuan Ma, Kang Wei, Long Shi, Ming Ding and Wen Chen.
  IEEE Journal on Selected Areas in Communications, vol. 41, no. 3, pp. 755-775, March 2023.
  (中科院1区, CCF-A)
  

• [TWC]  Blockchain Assisted Federated Learning over Wireless Channels: Dynamic Resource Allocation and Client Scheduling,
  Xiumei Deng, Jun Li, Chuan Ma, Kang Wei, Long Shi, Ming Ding, Wen Chen and H. Vincent Poor.
  Accepted by IEEE Transactions on Wireless Communications, 2022.
  (中科院1区, CCF-B)
  

• [IoT]  Federated Learning with Unreliable Clients: Performance Analysis and Mechanism Design,
  Chuan Ma, Jun Li, Ming Ding, Kang Wei, Wen Chen and H. Vincent Poor.
  IEEE Internet of Things Journal, vol. 8, no. 24, pp. 17308-17319, 2021.
  (中科院1区, CCF-C)
  

Books

• [B1]  Differentially Private Federated Learning: Algorithm, Analysis and Optimization,
  Kang Wei, Jun Li, Chuan Ma, Ming Ding and H. Vincent Poor.
  Federated Learning Systems, Springer Nature, 51-78, 2021.
  (Book Chapter)
  

Talks

• Invited Talk, “Communication Rounds Discounting in Federated Learning with Differential Privacy”, IEEE International Symposium on a Word of Wireless, Mobile and Multimedia Networks (WOWMOM)-Intelligent Things and Services (INTHINGS), Online, 2020.

• Tutorial, “Privacy and Security in Federated Learning”, IEEE/CIC ICCC, Online, 2020.

Academic Service

PC Member / Journal Reviewer

• Reviewer for:
  IEEE TIFS, IEEE JSAC, IEEE TSP, IEEE TKDE, IEEE TMC, IEEE TWC, IEEE JSTSP, IEEE TC, ACM SIGKDD, NeurIPS, IEEE ISIT and so on.

Selected Honors

• Excellent Academic Paper
  Awarded by Chinese Institute of Electronics (中国电子学会), 2023
• Post-doctoral Matching Fund
  Awarded by PolyU, 2023
• Wiley China Open Science Author of the Year
  Awarded by Wiley, 2022
• IEEE Signal Processing Society Best Paper Award
  Awarded by IEEE Signal Processing Society, 2022
• National Scholarship for Doctoral Students
  Awarded by Ministry of Education, 2021
• Tencent Rhinoceros Bird Project
  Awarded by Tencent, 2021
• Outstanding Doctoral Program
  Awarded by Nanjing University of Science and Technology, 2021-2023

Competitions

• First Place of iDASH Privacy and Security Workshop-Secure Genome Analysis Competition.
  Awarded by National Institutes of Health (美国国立卫生研究院), 2021

Patents

• 李骏, 韦康, 马川, 周程安, 梁腾. 一种针对联合学习中噪声攻击的防御方法. 中国发明专利, 专利号: ZL202011081206.3
• 孔令军, 李骏, 王诚, 韦康, 赵熙唯. 基于深度神经网络预测MLC闪存电压阈值的自适应方法. 中国发明专利, 专利号: ZL201810429797.5
• 赵熙唯, 吴平阳, 刘倩, 王诚, 李骏, 桂林卿, 韦康. 基于卷积神经网络降低信道译码误码率的方法. 中国发明专利, 专利号: ZL201910736687.8

Sponsors

I sincerely thank these sponsors for providing financial support for my research: