Chenjia Bai 白辰甲

I am a Researcher at Shanghai AI Laboratory. Prior to this, I obtained my Ph.D. degree in Computer Science from Harbin Institute of Technology (HIT), advised by Prof. Peng Liu. My research mainly focuses on deep Reinforcement Learning (RL), including offline RL, robust RL, efficient exploration, representation learning, risk-sensitive learning, and multi-agent RL.

I am fortunate to have been collaborated with many fantastic researchers. I was a visiting student at University of Toronto and Vector Institute, working with Prof. Animesh Garg . I was a visiting student at Northwestern University (remotely), working with Prof. Zhaoran Wang . I also used to be an intern at Huawei Noah's Ark Lab (advised by Prof. Jianye Hao), Tencent Robotics X (advised by Dr. Lei Han), and Alibaba. I received my Bachelor's degree and Master's degree in Computer Science from HIT.

Internship chances:
Our group is looking for highly-motivated Research Interns on board RL and Robotics. We are also interested in RL applications including Robot Arm and Quadruped. We have abundant GPU resources and robots to support your research.

Please drop me an email if you are interested in.

Publications

XXX for Offline Reinforcement Learning.
Chenjia Bai, Lingxiao Wang, Zhuoran Yang, Zhen Wang, Zhaoran Wang, and Xuelong Li
under review

SCORE: Spurious Correlation Reduction for Offline Reinforcement Learning. [arxiv]
Zhihong Deng, Zuyue Fu, Lingxiao Wang, Zhuoran Yang, Chenjia Bai, Zhaoran Wang, and Jing Jiang
under review

RORL: Robust Offline Reinforcement Learning via Conservative Smoothing. [arxiv]
Rui Yang*, Chenjia Bai*, Xiaoteng Ma, Zhaoran Wang, Chongjie Zhang, Lei Han
Neural Information Processing Systems (NeurIPS), 2022

Self-Supervised Imitation for Offline Reinforcement Learning with Hindsight Relabeling. [pdf]
Xudong Yu, Chenjia Bai*, Changhong Wang*, Zhen Wang
IEEE Transactions on Systems, Man, and Cybernetics: Systems (under review)

Contrastive UCB: Provably Efficient Contrastive Self-Supervised Learning in Online Reinforcement Learning. [arxiv, code]
Shuang Qiu, Lingxiao Wang, Chenjia Bai, Zhuoran Yang, and Zhaoran Wang
International Conference on Machine Learning (ICML), 2022

Monotonic Quantile Network for Worst-Case Offline Reinforcement Learning. [pdf]
Chenjia Bai, Ting Xiao, Zhoufan Zhu, Lingxiao Wang, Fan Zhou, Peng Liu, and Zhaoran Wang
IEEE Transactions on Neural Networks and Learning Systems, 2022

Exploration in Deep Reinforcement Learning: A Comprehensive Survey. [arxiv]
Tianpei Yang*, Hongyao Tang*, Chenjia Bai*, Jinyi Liu*, Jianye Hao, Zhaopeng Meng, and Peng Liu
(*Equally Contribution)
IEEE Transactions on Neural Networks and Learning Systems, 2022 (2nd review)

Pessimistic Bootstrapping for Uncertainty-Driven Offline Reinforcement Learning. [pdf, arxiv, code]
Chenjia Bai, Lingxiao Wang, Zhuoran Yang, Zhihong Deng, Animesh Garg, Peng Liu, and Zhaoran Wang
International Conference on Learning Representations (ICLR), 2022 Spotlight

OVD-Explorer: A General Information-theoretic Exploration Approach for Reinforcement Learning. [pdf]
Jinyi Liu, Wang Zhi, Yan Zheng, Jianye Hao, Junjie Ye, Chenjia Bai, Pengyi Li
NeurIPS Deep RL Workshop, 2021

Dynamic Bottleneck for Robust Self-Supervised Exploration. [arxiv, code]
Chenjia Bai, Lingxiao Wang, Lei Han, Animesh Garg, Jianye Hao, Peng Liu, and Zhaoran Wang
Neural Information Processing Systems (NeurIPS), 2021

Principled Exploration via Optimistic Bootstrapping and Backward Induction . [pdf, arxiv, code]
Chenjia Bai, Lingxiao Wang, Lei Han, Jianye Hao, Animesh Garg, Peng Liu, and Zhaoran Wang
International Conference on Machine Learning (ICML), 2021 Spotlight

Addressing Hindsight Bias in Multi-Goal Reinforcement Learning. [pdf, code]
Chenjia Bai, Lingxiao Wang, Yixin Wang, Zhaoran Wang, Rui Zhao, Chenyao Bai and Peng Liu
IEEE Transactions on Cybernetics, 2021

Variational Dynamic for Self-Supervised Exploration in Deep Reinforcement Learning. [arxiv, website]
Chenjia Bai, Peng Liu, Kaiyu Liu, Lingxiao Wang, Yingnan Zhao, Lei Han, and Zhaoran Wang
IEEE Transactions on Neural Networks and Learning Systems, 2021 .

Generating Attentive Goals for Prioritized Hindsight Reinforcement Learning. [pdf]
Peng Liu, Chenjia Bai, Yingnan Zhao, Chenyao Bai, Wei Zhao, and Xianglong Tang (supervisor first-author)
Knowledge-Based Systems (KBS), 2020

Obtaining Accurate Estimated Action Values in Categorical Distributional Reinforcement Learning. [pdf]
Yingnan Zhao, Peng Liu, Chenjia Bai, Wei Zhao, and Xianglong Tang
Knowledge-Based Systems (KBS), 2020

Active Sampling for Deep Q-learning Based on TD-error Adaptive Correction. [pdf]
Chenjia Bai, Peng Liu, Wei Zhao, and Xianglong Tang
Journal of Computer Research and Development (in Chinese), 2019.

Service

Conference Reviewer/Program Committee: NeurIPS (2021, 2022), ICLR (2021, 2022), ICML (2022), AAAI (2021, 2022)
Journal Reviewer: IEEE Trans. Cybernetics, IEEE Trans. TNNLS