Please feel free to email me c.shi7@lse.ac.uk if you have any comments.

Some Preprints

* indicates equal contribution

Zhang, J., Wang, J., Shi, C., Piette, J., Zeng, D., Wu, Z. PyCFRL: A Python library for counterfactually fair offline reinforcement learning via sequential data preprocessing

Wang, J., Wen, Q., Zhang, Y., Yan, X. and Shi, C. A Two-armed Bandit Framework for A/B Testing.

Ye, K*., Zhou, H*., Zhu, J*., Quinzan, F. and Shi, C. Robust Reinforcement Learning from Human Feedback for Large Language Models Fine-Tuning Python module VRPO

Zhu, J., Zhou, X., Yao, J., Aminian, G., Rivasplata, O., Little, S., Li, L. and Shi, C. Semi-pessimistic Reinforcement Learning.

Shen, G., Dai, R., Wu, G., Luo, S., Shi, C. and Zhu, H. Deep Distributional Learning with Non-crossing Quantile Network

Shi, C. Statistical Inference in Reinforcement Learning: A Selective Survey

Wang, J., Shi, C., Piette, J., Loftus, J., Zeng, D. and Wu, Z. Counterfactually Fair Reinforcement Learning via Sequential Data Preprocessing

Liu, P., Shi, C. and Sun, W. Dual Active Learning for Reinforcement Learning from Human Feedback

Sun, K., Kong, L., Zhu, H. and Shi, C. ARMA-Design: Optimal Treatment Allocation Strategies for A/B Testing in Partially Observable Time Series Experiments Python module ARMAdesign
slides

Hao, M*., Su, P*., Hu, L., Szabó, Z., Zhao, Q. and Shi, C. Off-policy Evaluation with Deeply-abstracted States . Python module state-abstraction

Dai, R*., Wang, J*., Zhou, F*., Luo, S., Qin, Q., Shi, C., and Zhu, H. Causal Deepsets for Off-policy Evaluation under Spatial or Spatio-temporal Interferences

Yang, Y., Shi, C., Yao, F., Wang, S. and Zhu, H. Spatially Randomized Designs Can Enhance Policy Evaluation

Wang, D., Shi, C., Luo, S. and Sun, W. Pessimistic Causal Reinforcement Learning with Mediators for Confounded Offline Data

Ma, T*., Zhu, J*., Cai, H., Qi, Z., Chen, Y., Shi, C. and Laber, E. Sequential Knockoffs for Variable Selection in Reinforcement Learning (SEEK)

Hu, L*., Li, M*., Shi, C., Wu, Z. and Fryzlewicz, P. Doubly Inhomogeneous Reinforcement Learning. Python module DIRL
slides presented at CMStatistics 2022.

Publications/accepted manuscripts