Publications

Check out my Google Scholar page here.

Preprints

  1. Robust Policy Gradient against Strong Data Corruption.
    Xuezhou Zhang, Yiding Chen, Xiaojin Zhu and Wen Sun.
    2021. [arXiv]

  2. Reward Poisoning in Reinforcement Learning: Attacks Against Unknown Learners in Unknown Environments.
    Amin Rakhsha*, Xuezhou Zhang*, Xiaojin Zhu and Adish Singla.
    2021.

  3. Using Machine Teaching to Investigate Human Assumptions when Teaching Reinforcement Learners.
    Xuezhou Zhang*, Yun-Shiuan Chuang*, Yuzhe Ma, Mark Ho, Joe Austerweil, Xiaojin Zhu.
    2020. [arXiv]

  4. Neural Additive Models:Interpretable Machine Learning with Neural Networks.
    Rishabh Agarwal, Nicholas Frosst, Xuezhou Zhang, Rich Caruana, Geoffrey Hinton.
    2020. [arXiv]

Conference Proceedings

  1. The sample complexity of teaching by reinforcement on Q-learning.
    Xuezhou Zhang, Shubham Kumar Bharti, Yuzhe Ma, Adish Singla, Xiaojin Zhu.
    AAAI 2021. [arXiv]

  2. Controllable and Diverse Text Generation in E-commerce.
    Huajie Shao, Jun Wang, Haohong Lin, Xuezhou Zhang, Aston Zhang, Heng Ji and Tarek Abdelzaher.
    WWW 2021.

  3. Task-agnostic Exploration in Reinforcement Learning.
    Xuezhou Zhang, Yuzhe Ma, Adish Singla.
    Neurips 2020. [arXiv]

  4. Adaptive Reward-Poisoning Attacks against Reinforcement Learning.
    Xuezhou Zhang, Yuzhe Ma, Adish Singla, Jerry Zhu.
    ICML 2020. [arXiv]

  5. Online Data Poisoning Attack.
    Xuezhou Zhang, Xiaojin Zhu, Laurent Lessard.
    L4DC 2020. Oral Presentation. [paper | poster]

  6. Policy poisoning in batch reinforcement learning and control.
    Yuzhe Ma, Xuezhou Zhang, Wen Sun, Xiaojin Zhu.
    NeurIPS 2019. [paper]

  7. Axiomatic Interpretability for Multiclass Additive Models.
    Xuezhou Zhang, Sarah Tan, Paul Koch, Urszula Chajewska, Rich Caruana.
    KDD 2019, Research Track. Oral Presentation. [paper]

  8. An Optimal Control Approach to Sequential Machine Teaching.
    Laurent Lessard*, Xuezhou Zhang*, Xiaojin Zhu.
    AISTATS 2019. [paper]

  9. Training Set Debugging using Trusted Items.
    Xuezhou Zhang, Xiaojin Zhu, Stephen J. Wright.
    AAAI 2018. Oral Presentation. [paper | slides | code]

  10. Teacher Improves Learning by Selecting a Training Subset.
    Yuzhe Ma, Robert D Nowak, Philippe Rigollet, Xuezhou Zhang, Xiaojin Zhu.
    AISTATS 2018. [paper | code]

  11. Training set camouflage.
    Ayon Sen, Scott Alfeld, Xuezhou Zhang, Ara Vartanian, Yuzhe Ma and Xiaojin Zhu.
    GameSec 2018. Oral Presentation. [paper]

  12. Optimal Teaching for Online Perceptrons.
    Xuezhou Zhang, Hrag Gorune Ohannessian, Ayon Sen, Scott Alfeld and Xiaojin Zhu.
    NIPS 2016 workshop on Constructive Machine Learning (CML). [paper | poster | slides]