Close

WU, Xiao-Ming

Assistant Professor, Computing@PolyU Hong Kong

News

About Me

I am currently an assistant professor in the Department of Computing, The Hong Kong Polytechnic University. I obtained my PhD degree from the Department of Electrical Engineering, Columbia University, advised by Prof. Shih-Fu Chang, with my thesis titled as "Learning on Graphs with Partially Absorbing Random Walks: Theory and Practice". My thesis research has been applied to Huawei App Store for large-scale App Push Recommendation, and significantly improved performance over previous methods.

Prior to that, I studied in the Department of Information Engineering, The Chinese University Of Hong Kong, and obtained a MPhil degree, advised by Prof. Shuo-Yen Robert Li and Prof. Anthony Man-Cho So. Before coming to Hong Kong, I studied in Peking University, where I received my BSc degree from the School of Mathematical Sciences, and my MSc degree from the Instituite of Computer Science and Technology, advised by Prof. Zongming Guo.

Research

My research interests are in machine learning, artificial intelligence, and data mining. I was drawn to the discipline of machine learning by the way it blends mathematics and real applications. I am mostly interested in developing unsupervised and semi-supervised learning algorithms and applying them to solve various real problems in computer vision, natural language processing, social network analysis, knowledge discovery and information retrieval, etc. On the theory side, I seek to develop deeper understanding into the principles of practical methods. For example, under what circumstance one method is better than another, and when a particular assumption breaks down. I usually end up gaining new insights by studying such questions, which give me inspirations of developing new and efficient methods. Some topics I am currently working on include:

Selected Publications

    * indicates corresponding author.

  • Contrastive Pre-training and Representation Distillation for Medical Visual Question Answering Based on Radiology Images
    Bo Liu, Li-Ming Zhan, Xiao-Ming Wu*
    To Appear in Proceedings of the 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), September 2021.
    [PDF] [Code]

  • Dimensionwise Separable 2-D Graph Convolution for Unsupervised and Semi-Supervised Learning on Graphs
    Qimai Li, Xiaotong Zhang, Han Liu, Quanyu Dai, Xiao-Ming Wu*
    To Appear in Proceedings of the 27th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD) (Research Track), August 2021.
    [PDF] [Code]

  • Embedding-based Product Retrieval in Taobao Search
    Sen Li, Fuyu Lv, Taiwei Jin, Guli Lin, Keping Yang, Xiaoyi Zeng, Xiao-Ming Wu, Qianli Ma
    To Appear in Proceedings of the 27th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD) (Applied Data Science Track), August 2021.
    [PDF] [Code]

  • Out-of-Scope Intent Detection with Self-Supervision and Discriminative Training
    Li-Ming Zhan, Haowen Liang, Bo LIU, Lu Fan, Xiao-Ming Wu*, Albert Y.S. Lam
    To Appear in Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL) (Long Paper), August 2021 (Oral).
    [PDF] [Code ]

  • SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answering
    Bo Liu, Li-Ming Zhan, Li Xu, Lin Ma, Yan Yang, Xiao-Ming Wu*
    To Appear in Proceedings of the 2021 IEEE International Symposium on Biomedical Imaging (ISBI), April 2021 (Oral).
    [PDF] [Dataset ]

  • A Closer Look at the Training Strategy for Modern Meta-Learning
    Jiaxin Chen, Xiao-Ming Wu*, Yanke Li, Qimai Li, Li-Ming Zhan, Fu-lai Chung*
    In Proceedings of the Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS), December 2020.
    [PDF] [Supplemental]

  • Medical Visual Question Answering via Conditional Reasoning
    Li-Ming Zhan, Bo Liu, Lu Fan, Jiaxin Chen, Xiao-Ming Wu*
    In Proceedings of the 28th ACM International Conference on Multimedia (ACM MM), October 2020.
    [PDF] [Code ]

  • M2GRL: A Multi-task Multi-view Graph Representation Learning Framework forWeb-scale Recommender Systems
    Menghan Wang*, Yujie Lin, Guli Lin, Keping Yang, Xiao-Ming Wu
    In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD) (Applied Data Science Track), August 2020 (Oral).
    [PDF] [Code]

  • Unknown Intent Detection Using Gaussian Mixture Model with an Application to Zero-shot Intent Classification
    Guangfeng Yan, Lu Fan, Qimai Li, Han Liu, Xiaotong Zhang, Xiao-Ming Wu*, Albert Y.S. Lam
    In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL) (Long Paper), July 2020.
    [PDF] [ Code ]

  • Variational Metric Scaling for Metric-Based Meta-Learning
    Jiaxin Chen, Li-Ming Zhan, Xiao-Ming Wu*, Fu-lai Chung*
    In Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI), February 2020 (Spotlight).
    [PDF] [Code]

  • Reconstructing Capsule Networks for Zero-shot Intent Classification
    Han Liu, Xiaotong Zhang, Lu Fan, Xuandi Fu, Qimai Li, Xiao-Ming Wu*, Albert Y.S. Lam
    In Proceedings of 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP) (Long Paper), November 2019.
    [PDF] [ Code ]

  • Attributed Graph Clustering via Adaptive Graph Convolution
    Xiaotong Zhang, Han Liu, Qimai Li, Xiao-Ming Wu*
    In Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI), August 2019.
    [PDF] [Code]

  • Label Efficient Semi-Supervised Learning via Graph Filtering
    Qimai Li, Xiao-Ming Wu*, Han Liu, Xiaotong Zhang, Zhichao Guan
    In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2019.
    [PDF] [Code]

  • Large Margin Meta-Learning for Few-Shot Classification
    Yong Wang, Xiao-Ming Wu*, Qimai Li, Jiatao Gu, Wangmeng Xiang, Lei Zhang, Victor O.K.Li*
    In Thirty-second Annual Conference on Neural Information Processing Systems Workshop ( NeurIPSW) on Meta-Learning, December 2018.
    [Workshop version] [Early long version on arXiv] [Code]

  • Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning
    Qimai Li, Zhichao Han, Xiao-Ming Wu*
    In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence (AAAI), February 2018 (Oral).
    [PDF] [Project] [Code]

  • Chapter 14: Partially Absorbing Random Walks: A Unified Framework for Learning on Graphs
    Xiao-Ming Wu*, Zhenguo Li, and Shih-Fu Chang.
    Book Chapter in Cooperative and Graph Signal Processing -- Principles and Applications, Elsevier, June 2018.

  • New Insights into Laplacain Similarity Search.
    Xiao-Ming Wu, Zhenguo Li, and Shih-Fu Chang.
    In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2015
    [PDF] [Supplemental] [Abstract] [Code] [Poster]

  • Locally Linear Hashing for Extracting Non-Linear Manifolds.
    Go Irie, Zhenguo Li, Xiao-Ming Wu, and Shih-Fu Chang.
    In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2014.
    [PDF] [Supplemental] [Code] [Poster]

  • Analyzing the Harmonic Structure in Graph-Based Learning.
    Xiao-Ming Wu, Zhenguo Li, and Shih-Fu Chang.
    In Proceedings of Advances in Neural Information Processing Systems (NeurIPS), December 2013.
    [PDF] [Supplemental] [Code] [Poster]

  • Learning with Partially Absorbing Random Walks.
    Xiao-Ming Wu, Zhenguo Li, Anthony Man-Cho So, John Wright, and Shih-Fu Chang.
    In Proceedings of Advances in Neural Information Processing Systems (NeurIPS), December 2012.
    [PDF] [Supplemental] [Code] [Poster]

  • Segmentation Using Superpixels: A Bipartite Graph Partitioning Approach.
    Zhenguo Li, Xiao-Ming Wu, and Shih-Fu Chang.
    In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2012.
    [PDF] [Code] [Project][Poster]

  • Fast Graph Laplacian Regularized Kernel Learning via Semidefinite-Quadratic-Linear Programming.
    Xiao-Ming Wu, Anthony Man-Cho So, Zhenguo Li, and Shuo-Yen Robert Li.
    In Proceedings of Advances in Neural Information Processing Systems (NeurIPS), December 2009 (Spotlight).
    [PDF] [Code] [Poster]

My Group

I am lucky to work with brilliant students who are willing to follow my research interests, and tackle research problems with me.

Teaching Subjects

I have enjoyed teaching the following subjects.

Contact

The best way to reach me is by email. Due to the large volume of emails recieved, I could not respond to every enquiry, but I do read every email. If I do not respond to your inquiry about PhD/MPhil/RA position, it does not necessarily mean you are not suitable for it but may be simply that it is not available.