state-action pairs

3 papers

• Deployment Efficient Reward-Free Exploration with Linear Function Approximation
• OpenCUA: Open Foundations for Computer-Use Agents
• STAR: Efficient Preference-based Reinforcement Learning via Dual Regularization