Fengxiang He

PhD student
UBTECH Sydney AI Centre
School of Computer Science
Faculty of Engineering
University of Sydney

Brief bio

I received my BSc in statistics from University of Science and Technology of China, and my MPhil (Engineering and IT) from the University of Sydney, supervised by Professor Dacheng Tao FAA, where I am pursuing a PhD (Engineering and IT) with the same supervisor. I was research assistant at Nanyang Technological University. My research interest is in theoretical foudnations of machine learning (especially deep learning) and their applications in designing accountable and responsible machine learning algorithms.


See also at my Google Scholar, DBLP, and Semantic Scholar profiles. Any comment is welcomed.


  1. Fengxiang He and Dacheng Tao. "Recent advances in deep learning theory." 2020. [paper] [bib]

Research papers

  1. Shaopeng Fu*, Fengxiang He*, and Dacheng Tao. "Bayesian inference forgetting." 2021. [paper] [code] [bib]

  2. Fengxiang He*, Shiye Lei*, Jianmin Ji, and Dacheng Tao. "Neural networks behave as hash encoders: An empirical study." 2021. [paper] [code] [bib]

  3. Fengxiang He*, Shaopeng Fu*, Bohan Wang*, and Dacheng Tao. "Robustness, privacy, and generalization of adversarial training." 2020. [paper] [code] [bib]

  4. Fengxiang He*, Bohan Wang*, and Dacheng Tao. "Tighter generalization bounds for iterative differentially private learning algorithms." 2020. [paper] [bib]

  5. Fengxiang He*, Bohan Wang*, and Dacheng Tao. "Piecewise linear activations substantially shape the loss surfaces of neural networks." ICLR 2020. [paper] [website] [poster] [bib]

  6. Zhuozhuo Tu, Fengxiang He, and Dacheng Tao. "Understanding generalization in recurrent neural networks." ICLR 2020. [paper] [website] [bib]

  7. Fengxiang He, Tongliang Liu, and Dacheng Tao. "Control batch size and learning rate to generalize well: Theoretical and empirical evidence." NeurIPS 2019. [paper] [poster] [bib]

  8. Fusheng Hao, Fengxiang He, Jun Cheng, Lei Wang, Jianzhong Cao, and Dacheng Tao. "Collect and select: Semantic alignment metric learning for few-shot learning." ICCV 2019. [paper] [code] [bib]

  9. Jingyuan Li, Fengxiang He, Lefei Zhang, Bo Du, Yonghao Xu, and Dacheng Tao. "Progressive reconstruction of visual structure for image inpainting." ICCV 2019. [paper] [code] [bib]

  10. Fengxiang He, Tongliang Liu, and Dacheng Tao. "Why ResNet works? Residuals generalize." TNNLS. 2020. [paper] [bib]

  11. Sheng Li, Fengxiang He, Bo Du, Lefei Zhang, Yonghao Xu, Dacheng Tao. "Fast spatio-temporal residual network for video super-resolution." CVPR 2019. [paper] [bib]

* Co-first authors.



Level 1 32 Queen St
+61 402 167 965

Last update: Sat 28 Jan 2021