LS-IQ: Implicit Reward Regularization for Inverse Reinforcement Learning
This work addresses stability issues in imitation learning for robotics or autonomous systems, but it is incremental as it builds on existing implicit reward methods.
The paper tackled the problem of instability and mistreatment of absorbing states in inverse reinforcement learning by proposing LS-IQ, a method that uses implicit reward regularization to minimize a bounded χ²-divergence, resulting in outperforming state-of-the-art algorithms, especially in environments with absorbing states.
Recent methods for imitation learning directly learn a $Q$-function using an implicit reward formulation rather than an explicit reward function. However, these methods generally require implicit reward regularization to improve stability and often mistreat absorbing states. Previous works show that a squared norm regularization on the implicit reward function is effective, but do not provide a theoretical analysis of the resulting properties of the algorithms. In this work, we show that using this regularizer under a mixture distribution of the policy and the expert provides a particularly illuminating perspective: the original objective can be understood as squared Bellman error minimization, and the corresponding optimization problem minimizes a bounded $χ^2$-Divergence between the expert and the mixture distribution. This perspective allows us to address instabilities and properly treat absorbing states. We show that our method, Least Squares Inverse Q-Learning (LS-IQ), outperforms state-of-the-art algorithms, particularly in environments with absorbing states. Finally, we propose to use an inverse dynamics model to learn from observations only. Using this approach, we retain performance in settings where no expert actions are available.