Chengchun Shi

(i), I am an associate professor of data science at London School of Economics and Political Science (LSE).

(ii) Pior to (i), I was an assistant professor of data science at LSE.

(iii) Prior to (ii), I was a Ph.D. student in Statistics at North Carolina State University (NCSU). I work with Dr. Wenbin Lu and Dr. Rui Song.

(iv) Prior to (iii), I obtained a B.S. in Statistics from Zhejiang University in July 2014.

(v) Prior to (iv), I graduated from Ningbo Xiaoshi High School in July 2010.

I am looking for PhD students interested in reinforcement learning (see my lecture slides) and LLMs.

August 2025: Talk @ JSM on Doubly Robust LLM Fine-Tuning.
July 2025: RL short course at Renmin University, Shandong University, and Capital Normal University (Slides).
July 2025: Talk @ Tsinghua Statistics + AI Frontier Summit, JCSDS, SUFE.
June 29, 2025: Talk @ ICSA, Zhuhai, China.
- Erhan will present our paper on Doubly Robust LLM Fine-Tuning on my behalf (Slides)
- Jin will present our work on LLM detection
June 21, 2025: Hongyi will present our work on LLM detection @ 狗熊会
May 2025: Four papers accepted to ICML 2025. Congratulations to all co-authors! Talk @ Warwick
April 2025: Honored to receive the Peter Gavin Hall IMS Early Career Prize.

My research is motivated from the following applications:

LLMs (see our recent papers DRPO, VRPO, slides on fine-tuning);
Ridesharing (see our AAAI tutorial (Youtube, Bilibili) and my talk; see also some simulated environments for Order Dispatch and Spatio-temporal Policy Evaluation);
Video-sharing (see our KDD paper for details about our proposal successfully deployed in a widely used mobile app with millions of daily active users)
Mobile health (some simulated environments for Diabetes and Intern Health);
Neuroscience (see our paper on using RL for modelling human decision making)
Precision medicine (a simulated STARD data example).

Some of my recent talks and slides on statistical inference, RL, causal inference, data integration, time series and experimental design:

StatRL Bilibili, Youtube; Chinese versions: Bilibili, Youtube; slides; the accompanying paper and its Chinese version
CausalRL Youtube, slides; the accompanying paper
ARMAdesign 2-hour slides, 1-hour slides, 30-minute slides; the accompanying paper
OPE my slides and our review paper;
Pessimistic Data Integration slides; the accompanying paper
Doubly Inhomogeneous policy learning, policy evaluation.

A summary of my past research: Statistical Methods in Reinforcement Learning (partly supported by EPSRC)