吴 锦 辉 Jin-Hui Wu Lecturer School of Computer Science and Technology Soochow University, Suzhou, Jiangsu, China Email: wujinhui@suda.edu.cn

Biography

• Oct 2025 - Present: Lecturer, School of Computer Science and Technology, Soochow University.
• Sep 2019 - Sep 2025: Ph.D., School of Artificial Intelligence, Nanjing University. (without entrance examination)
  Supervisors: Prof. Yuan Jiang and Prof. Zhi-Hua Zhou.
• Sep 2015 - Jun 2019: B.Sc., Kuang Yaming Honors School, Nanjing University.
  Supervisor: Prof. Wei Wang.

Publications

• Reparameterized Complex-valued Neurons Can Efficiently Learn More than Real-valued Neurons via Gradient Descent. [paper] [bib]
  Jin-Hui Wu, Shao-Qun Zhang, Yuan Jiang, Zhi-Hua Zhou.
  Journal of Machine Learning Research (JMLR), 27(38): 1-51, 2026.


• Universal Approximation and Approximation Advantages of Quaternion-valued Neural Networks [in Chinese]. [paper] [bib]
  Jin-Hui Wu, Yuan Jiang.
  Journal of Computer Research and Development, 62(5): 1205-1215, 2025.
  (Best Student Paper Award of the 10th China Conference on Data Mining (CCDM), 2024)


• The Role of Depth, Width, and Tree Size in Expressiveness of Deep Forest. [paper] [code] [bib]
  Shen-Huan Lyu*, Jin-Hui Wu*, Qin-Cheng Zheng, Baoliu Ye.
  In: Proceedings of the 27th European Conference on Artificial Intelligence (ECAI), pp. 2042-2049, 2024.


• On the Intrinsic Structures of Spiking Neural Networks. [paper] [bib]
  Shao-Qun Zhang, Jia-Yi Chen, Jin-Hui Wu, Gao Zhang, Huan Xiong, Bin Gu, Zhi-Hua Zhou.
  Journal of Machine Learning Research (JMLR), 25(194): 1-74, 2024.


• Theoretical Exploration of Flexible Transmitter Model. [arXiv] [TNNLS] [bib]
  Jin-Hui Wu, Shao-Qun Zhang, Yuan Jiang, Zhi-Hua Zhou.
  IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 35(3): 3674-3688, 2024.


• Complex-valued Neurons Can Learn More but Slower than Real-valued Neurons via Gradient Descent. [paper] [poster] [code] [bib] [corrigenda]
  Jin-Hui Wu, Shao-Qun Zhang, Yuan Jiang, Zhi-Hua Zhou.
  In: Advances in Neural Information Processing Systems 36 (NeurIPS), pp. 23714-23747, 2023.

Links

Correspondence