I am currently a Ph.D. student working in School of Computing, National University of Singapore, supervised by Prof. Dong Jin Song.

My research interests include trustworthy AI, graph neural networks and many other computer science topics.

I got my Master's and B.Eng. degree in NUS and Xidian University respectively.

RESEARCH πŸ§ͺ

coming soon...

Google Scholar | dblp | GitHub | Linkedin | Twitter

PUBLICATIONS 🧻 [view full list]


  • 2024
  • Xinke Li, Junchi Lu, Henghui Ding Changsheng Sun, Joey Tianyi Zhou, Chee Yeow Meng.
    PointCVaR: Risk-optimized Outlier Removal for Robust 3D Point Cloud Classification. Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI), 2024. [pdf | website | code]


  • 2022
  • Yan Xiao, Yun Lin, Ivan Beschastnikh, Changsheng Sun, David S. Rosenblum, Jin Song Dong.
    Repairing Failure-inducing Inputs with Input Reflection. The 37th IEEE/ACM International Conference on Automated Software Engineering (ASE) 2022. [pdf | code]

  • Hui Li, Mengting Xu, Sourav S Bhowmick, Joty Shafiq Rayhan, Changsheng Sun, Jiangtao Cui.
    PIANO: Influence Maximization Meets Deep Reinforcement Learning. IEEE Transactions on Computational Social Systems. [pdf | ieee]

    Since its introduction in 2003, the influence maximization (IM) problem has drawn significant research attention in the literature. The aim of IM, which is NP-hard, is to select a set of k users known as seed users who can influence the most individuals in the social network. The state-of-the-art algorithms estimate the expected influence of nodes based on sampled diffusion paths. As the number of required samples has been recently proven to be lower bounded by a particular threshold that presets tradeoff between the accuracy and the efficiency, the result quality of these traditional solutions is hard to be further improved without sacrificing efficiency. In this article, we present an orthogonal and novel paradigm to address the IM problem by leveraging deep reinforcement learning (RL) to estimate the expected influence. In particular, we present a novel framework called deeP reInforcement leArning-based iNfluence …


  • 2021
  • Yan Xiao, Ivan Beschastnikh, David S. Rosenblum, Changsheng Sun, Sebastian Elbaum, Y. Lin, Jin Song Dong.
    Self-Checking Deep Neural Networks in Deployment. The 43rd IEEE/ACM International Conference on Software Engineering (ICSE), 2021.. [pdf | code]

    ABSTRACTPaper(PDF)


  • 2020
  • Zekun Tong, Yuxuan Liang, Changsheng Sun, Xinke Li, David S. Rosenblum, Andrew Lim.
    Digraph Inception Convolutional Networks. Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS), 2020.. [pdf | poster | code]


  • preprints
  • Yan Xiao, Yun Lin, Ivan Beschastnikh, Changsheng Sun, David S Rosenblum, Jin Song Dong
    Generalizing Neural Networks by Reflecting Deviating Data in Production arXiv:2110.02718. [arXiv | pdf]

    ABSTRACTPaper(PDF)

  • Zekun Tong, Yuxuan Liang, Changsheng Sun, David S. Rosenblum, Andrew Lim.
    Directed graph convolutional network. arXiv:2004.13970. [arXiv | pdf]

    ABSTRACTPaper(PDF)

  • Hui Li, Mengting Xu, Sourav S Bhowmick, Changsheng Sun, Zhongyuan Jiang, Jiangtao Cui
    Disco: Influence maximization meets network embedding and deep learning. arXiv:1906.07378.. [arXiv | pdf]

    Since its introduction in 2003, the influence maximization (IM) problem has drawn significant research attention in the literature. The aim of IM is to select a set of k users who can influence the most individuals in the social network. The problem is proven to be NP-hard. A large number of approximate algorithms have been proposed to address this problem. The state-of-the-art algorithms estimate the expected influence of nodes based on sampled diffusion paths. As the number of required samples have been recently proven to be lower bounded by a particular threshold that presets tradeoff between the accuracy and efficiency, the result quality of these traditional solutions is hard to be further improved without sacrificing efficiency. In this paper, we present an orthogonal and novel paradigm to address the IM problem by leveraging deep learning models to estimate the expected influence. Specifically, we present a novel framework called DISCO that incorporates network embedding and deep reinforcement learning techniques to address this problem. Experimental study on real-world networks demonstrates that DISCO achieves the best performance w.r.t efficiency and influence spread quality compared to state-of-the-art classical solutions. Besides, we also show that the learning model exhibits good generality.

TEACHING πŸŽ“

BIO πŸ›Έ

PhD Candidate @NUS (supervisor: Dong Jin Song)◄ MComp @NUS◄ Research Intern & Algorithm Engineer @ JD Intelligent Cities Research , China. (2018) ◄ BEng @Xidian University.

CONTACT πŸ“

Sun Changsheng (ε­™εΈΈζ™Ÿ)
Programming Languages & Software Engineering Lab (PLSE@NUS),
Computing 3 #02-20,
11 Research Link, Singapore 119391.

Email: cssun[at]u[dot]nus[dot]edu

Β© Sun Changsheng | Updated 2023-12-16