Chenjia Bai 白辰甲
Hi, this is Chenjia Bai. I'm a Ph.D. student in Computer Science at Harbin Institute of Technology (HIT), advised by Prof. Peng Liu. I was a visiting student at University of Toronto and Vector Institute, working with Prof. Animesh Garg . My research mainly focuses on deep Reinforcement Learning (RL), including offline RL, robust RL, efficient exploration, representation learning, risk-sensitive learning, and multi-goal RL.
Before that, 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.
Publications
-
RORL: Robust Offline Reinforcement Learning via Conservative Smoothing. [arxiv]
Rui Yang*, Chenjia Bai*, Xiaoteng Ma, Zhaoran Wang, Chongjie Zhang, Lei Han
under review
-
Hindsight Self-Supervision for Offline Reinforcement Learning. [pdf]
Xudong Yu, Chenjia Bai, Changhong Wang, Zhen Wang
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
-
Contrastive UCB: Provably Efficient Contrastive Self-Supervised Learning in Online Reinforcement Learning. [arxiv]
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, Zhoufan Zhu, Ting Xiao, Lingxiao Wang, Fan Zhou, Peng Liu, and Zhaoran Wang
IEEE Transactions on Neural Networks and Learning Systems, 2022 (2nd review)
-
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 (under 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
-
Guided Goal Generation for Hindsight Multi-goal Reinforcement Learning. [pdf]
Chenjia Bai, Peng Liu, Wei Zhao, and Xianglong Tang
Neurocomputing, 2019
-
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
© 2022 Chenjia Bai