Learning Interpretable Policies in Hindsight-Observable POMDPs through Partially Supervised Reinforcement Learning

arXiv:2402.09290v1 Announce Type: new
Abstract: Deep reinforcement learning has demonstrated remarkable achievements across diverse domains such as video games, robotic control, autonomous driving, and drug discovery. Common methodologies in partially-observable domains largely lean on end-to-end learning from high-dimensional observations, such as images, without explicitly reasoning about true state. We suggest an alternative direction, introducing the Partially Supervised Reinforcement Learning (PSRL) framework. At the heart of PSRL is the fusion of both supervised and unsupervised learning. The approach leverages a state …

abstract arxiv autonomous autonomous driving control cs.ai cs.lg discovery diverse domains driving drug discovery games images lean observable reasoning reinforcement reinforcement learning robotic state through true type video video games