Switching the Loss Reduces the Cost in Batch (Offline) Reinforcement Learning
Machine Learning
2024-08-02 v5
Abstract
We propose training fitted Q-iteration with log-loss (FQI-log) for batch reinforcement learning (RL). We show that the number of samples needed to learn a near-optimal policy with FQI-log scales with the accumulated cost of the optimal policy, which is zero in problems where acting optimally achieves the goal and incurs no cost. In doing so, we provide a general framework for proving small-cost bounds, i.e. bounds that scale with the optimal achievable cost, in batch RL. Moreover, we empirically verify that FQI-log uses fewer samples than FQI trained with squared loss on problems where the optimal policy reliably achieves the goal.
Keywords
Cite
@article{arxiv.2403.05385,
title = {Switching the Loss Reduces the Cost in Batch (Offline) Reinforcement Learning},
author = {Alex Ayoub and Kaiwen Wang and Vincent Liu and Samuel Robertson and James McInerney and Dawen Liang and Nathan Kallus and Csaba Szepesvári},
journal= {arXiv preprint arXiv:2403.05385},
year = {2024}
}