递归熵风险强化学习的近最优样本复杂度

1. 12:00
   arXiv · Statistics Machine Learning

   递归熵风险强化学习在生成模型下的近最优样本复杂度

   该论文研究有限折扣 MDP 中递归熵风险偏好下的价值与策略学习样本复杂度，风险参数为 β≠0，并假设可访问 MDP 的生成模型。作者对基于模型的 risk-sensitive Q-value iteration（MB-RS-QVI）给出细化分析，为最优 Q 值函数与 ε-最优策略导出 (ε,δ)-PAC 保证。

Current heat 9·Comparable peak 10（Oct 7）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.