Policy Finetuning in Reinforcement Learning via Design of Experiments using Offline Data | HTML5

In some applications of reinforcement learning, a dataset of pre-collected experience is already available but it is also possible to acquire some additional online data to help improve the quality of the policy. Howev…

[2307.04354] Policy Finetuning in Reinforcement Learning via Design of Experiments using Offline Data Policy Finetuning in Reinforcement Learning via Design of Experiments using Offline Data Ruiqi Zhang Department of Statistics University of California Berkeley rqzhang@berkeley.edu Andrea Zanette Department of EECS University of California Berkeley zanette@berkeley.edu Abstract In some applications of reinforcement learning, a dataset of pre-collected experience is already available but it is also possible to acquire some additional online data to help improve the quality of the policy. Howeve

related reading

• Exploration for the Efficient Deployment of Reinforcement Learning Agentsopenreview.net
• [2103.06326] S4RL: Surprisingly Simple Self-Supervision for Offline Reinforcement Learningarxiv.org
• Reinforcement Learning in Newcomblike Problemsproceedings.neurips.cc
• [1707.06347] Proximal Policy Optimization Algorithmsarxiv.org
• 2109.10813.pdfarxiv.org
• [1709.06560] Deep Reinforcement Learning that Mattersarxiv.org
• [2005.01643] Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemsar5iv.labs.arxiv.org
• c21f4ce780c5c9d774f79841b81fdc6d-Paper.pdfproceedings.neurips.cc
• [2506.22401] Exploration from a Primal-Dual Lens: Value-Incentivized Actor-Critic Methods for Sample-Efficient Online RLarxiv.org
• RL_Notes__final_.pdfjubayer-ibn-hamid.github.io
• Value estimation with finite datamcgill.scholaris.ca
• [2306.01157] Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confoundingarxiv.org