Importance of Empirical Sample Complexity Analysis for Offline Reinforcement Learning
This work addresses the need for empirical sample complexity analysis to guide practical applications of offline RL, but it is incremental as it focuses on evaluation rather than new algorithms.
The paper tackles the problem of understanding sample complexity in offline reinforcement learning by proposing an evaluation approach to analyze how learning depends on the number of samples, aiming to assess the usefulness of existing algorithms.
We hypothesize that empirically studying the sample complexity of offline reinforcement learning (RL) is crucial for the practical applications of RL in the real world. Several recent works have demonstrated the ability to learn policies directly from offline data. In this work, we ask the question of the dependency on the number of samples for learning from offline data. Our objective is to emphasize that studying sample complexity for offline RL is important, and is an indicator of the usefulness of existing offline algorithms. We propose an evaluation approach for sample complexity analysis of offline RL.