Aldo Pacchiano

In-Context Learning for Pure Exploration

Post-training Large Language Models for Diverse High-Quality Responses

Select the Right Agent: Data-Driven Online Model Selection in Reinforcement Learning

Learning with Coupled Uncertainty

Learning to Undo: Transfer Reinforcement Learning under State Space Transformations

Principled Fine-tuning of LLMs from User-Edits: A Medley of Preference, Supervision, and Reward

Feasible Action Search for Bandit Linear Programs via Thompson Sampling

A Theoretical Framework for Partially-Observed Reward States in RLHF

Second Order Bounds for Contextual Bandits with Function Approximation

High Probability Contextual Bandits for Optimal Dosage Selection

Language Model Personalization via Reward Factorization

ORSO: Accelerating Reward Design via Online Reward Selection and Policy Optimization

Multiple-policy Evaluation via Density Estimation

Adaptive Exploration for Multi-Reward Multi-Policy Evaluation

Sample Efficient Multiple-policy Evaluation in Reinforcement Learning