Xiangxiang Xu(徐祥祥)


Ph.D.

Research Assistant, Tsinghua-Berkeley Shenzhen Institute

CV, Google Scholar

Email: xiangxiangxu AT ieee.org


My research focuses on the information-theoretic understanding of machine learning algorithms, together with its applications in data analytics.

Recent works

  1. Shao-Lun Huang, Xiangxiang Xu, and Lizhong Zheng. An information-theoretic approach to unsupervised feature selection for high-dimensional data". In IEEE Journal on Selected Areas in Information Theory (2020).
  2. Xiangxiang Xu and Shao-Lun Huang. Maximal Correlation Regression. In: IEEE Access 8 (2020), pp. 26591-26601. [Paper]
  3. Shao-Lun Huang, Xiangxiang Xu, Lizhong Zheng, and Gregory W. Wornell. An Information Theoretic Interpretation to Deep Neural Networks. In IEEE International Symposium on Information Theory (ISIT), 2019. [Paper] [arXiv]
  4. Lichen Wang, Jiaxiang Wu, Shao-Lun Huang, Lizhong Zheng, Xiangxiang Xu, Lin Zhang, and Junzhou Huang. An Efficient Approach to Informative Feature Extraction from Multimodal Data. In AAAI 2019. [Paper] [arXiv]
  5. Xiangxiang Xu, Shao-Lun Huang, Lizhong Zheng, and Lin Zhang. The Geometric Structure of Generalized Softmax Learning. In 2018 IEEE Information Theory Workshop (ITW) (IEEE ITW 2018), Guangzhou, P.R. China, November 2018. [Paper] [Slides]