Unified Algorithms for RL with Decision-Estimation Coefficients: No-Regret, PAC, and Reward-Free Learning
Unified Algorithms for RL with Decision-Estimation Coefficients: No-Regret, PAC, and Reward-Free Learning
Fan Chen,Song Mei,Yu Bai
2022 · DOI: 10.48550/arXiv.2209.11745
arXiv.org · 33 Citations
TLDR
Two new DEC-type complexity measures are proposed: Explorative DEC (EDEC), and Reward-Free DEC (RFDEC), and it is shown that they are necessary and sufficient for sample-efficient PAC learning and reward-free learning, thereby extending the original DEC which only captures no-regret learning.