Combining Experimental and Historical Data for Policy Evaluation
報告人簡介
史成春博士，現任倫敦政治經濟學院統計系副教授，曾在北卡羅來納州立大學（North Carolina State University）獲得統計學博士學位。他的研究主要集中在強化學習領域（Reinforcement Learning），特別是在策略評估（Policy Evaluation）、因果推斷（Causal Inference）、半監督學習（Semi-Supervised Learning）等方面的應用與優化。史博士曾榮獲Institute of Mathematical Statistics (IMS) Tweedie Award和Royal Statistical Society (RSS) Research Prize等獎項。
內容簡介
This talk considers policy evaluation with multiple data sources, especially in scenarios that involve one experimental dataset with two arms, complemented by a historical dataset generated under a single control arm. We propose novel data integration methods that linearly integrate base policy value estimators constructed based on the experimental and historical data, with weights optimized to minimize the mean square error (MSE) of the resulting combined estimator. We further apply the pessimistic principle to obtain more robust estimators, and extend these developments to sequential decision making. Theoretically, we establish non-asymptotic error bounds for the MSEs of our proposed estimators, and derive their oracle, efficiency and robustness properties across a broad spectrum of reward shift scenarios. Numerical experiments and real-data-based analyses from a ridesharing company demonstrate the superior performance of the proposed estimators.
