ベルマン勾配反復による逆強化学習
Bellman Gradient Iteration for Inverse Reinforcement Learning
観測された行動から報酬関数を復元する逆強化学習手法を提案し、Q値と報酬の微分可能な関係を構築して勾配法で学習する。
著者: Kun Li, Yanan Sui, Joel W. Burdick
分類: cs.LG, cs.RO
原文アブストラクト
This paper develops an inverse reinforcement learning algorithm aimed at recovering a reward function from the observed actions of an agent. We introduce a strategy to flexibly handle different types of actions with two approximations of the Bellman Optimality Equation, and a Bellman Gradient Iteration method to compute the gradient of the Q-value with respect to the reward function. These methods allow us to build a differentiable relation between the Q-value and the reward function and learn an approximately optimal reward function with gradient methods. We test the proposed method in two simulated environments by evaluating the accuracy of different approximations and comparing the proposed method with existing solutions. The results show that even with a linear reward function, the proposed method has a comparable accuracy with the state-of-the-art method adopting a non-linear reward function, and the proposed method is more flexible because it is defined on observed actions instead of trajectories.
