A Novel Approach to Curiosity and Explainable Reinforcement Learning via Interpretable Sub-Goals

Two key challenges within Reinforcement Learning involve improving (a) agent learning within environments with sparse extrinsic rewards and (b) the explainability of agent actions. We describe a curious subgoal focused agent to address both these challenges. We use a novel method for curiosity produced from a Generative Adversarial Network (GAN) based model of environment transitions that is robust to stochastic environment transitions. Additionally, we use a subgoal generating network to guide navigation. The explainability of the agent's behavior is increased by decomposing complex tasks into a sequence of interpretable subgoals that do not require any manual design. We show that this method also enables the agent to solve challenging procedurally-generated tasks that contain stochastic transitions above other state-of-the-art methods.