Subhojyoti Mukherjee

Optimal Design for Human Feedback

Speed: Experimental design for policy evaluation in linear heteroscedastic bandits

Experimental Design for Active Transductive Inference in Large Language Models

Multi-task Representation Learning for Pure Exploration in Bilinear Bandits

Efficient and Interpretable Bandit Algorithms

Revar: Strengthening policy evaluation via reduced variance sampling

Safety Aware Changepoint Detection for Piecewise iid Bandits

Chernoff sampling for active testing and extension to active regression