9 papers
Learning Perturbations to Extrapolate Your LLM
Zetai Cen, Chenfei Gu, Jin Zhu +3
Recent advancements in large language models demonstrate that injecting perturbations can substantially enhance extrapolation performance. However, current approaches often rely on…
Robust Sequential Experimental Design for A/B Testing
Qianglin Wen, Xiangkun Wu, Chengchun Shi +4
Experimental design has emerged as a powerful approach for improving the sample efficiency of A/B testing, yet existing designs rely critically on correctly specified models. We st…
Designing Time Series Experiments in A/B Testing with Transformer Reinforcement Learning
Xiangkun Wu, Qianglin Wen, Yingying Zhang +3
A/B testing has become a gold standard for modern technological companies to conduct policy evaluation. Yet, its application to time series experiments, where policies are sequenti…
Spatially Randomized Designs Can Enhance Policy Evaluation
Ying Yang, Chengchun Shi, Fang Yao +2
This article studies the benefits of using spatially randomized experimental designs which partition the experimental area into distinct, non-overlapping units with treatments assi…
Unraveling the Interplay between Carryover Effects and Reward Autocorrelations in Switchback Experiments
Qianglin Wen, Chengchun Shi, Ying Yang +2
A/B testing has become the gold standard for policy evaluation in modern technological industries. Motivated by the widespread use of switchback experiments in A/B testing, this pa…
A Two-armed Bandit Framework for A/B Testing
Jinjuan Wang, Qianglin Wen, Yu Zhang +2
A/B testing is widely used in modern technology companies for policy evaluation and product deployment, with the goal of comparing the outcomes under a newly-developed policy again…