Cyclophobic Reinforcement Learning
This addresses exploration challenges in reinforcement learning for complex environments, though it appears incremental as it builds on existing intrinsic reward methods.
The paper tackles the problem of sparse rewards in reinforcement learning by proposing a cyclophobic intrinsic reward that punishes redundancy to avoid cycles, achieving more sample-efficient results than state-of-the-art methods in MiniGrid and MiniHack environments.
In environments with sparse rewards, finding a good inductive bias for exploration is crucial to the agent's success. However, there are two competing goals: novelty search and systematic exploration. While existing approaches such as curiosity-driven exploration find novelty, they sometimes do not systematically explore the whole state space, akin to depth-first-search vs breadth-first-search. In this paper, we propose a new intrinsic reward that is cyclophobic, i.e., it does not reward novelty, but punishes redundancy by avoiding cycles. Augmenting the cyclophobic intrinsic reward with a sequence of hierarchical representations based on the agent's cropped observations we are able to achieve excellent results in the MiniGrid and MiniHack environments. Both are particularly hard, as they require complex interactions with different objects in order to be solved. Detailed comparisons with previous approaches and thorough ablation studies show that our newly proposed cyclophobic reinforcement learning is more sample efficient than other state of the art methods in a variety of tasks.