Exploring through Random Curiosity with General Value Functions
This addresses the problem of exploration inefficiency in reinforcement learning for researchers and practitioners, offering a novel approach that is incremental over existing curiosity-based methods.
The paper tackles the challenge of efficient exploration in reinforcement learning, particularly in partially observable environments, by proposing RC-GVF, an intrinsic reward function based on predicting general value functions. The result shows improved exploration in a hard-exploration problem and outperforms previous methods in MiniGrid environments, achieving competitive performance with baselines using privileged information.
Efficient exploration in reinforcement learning is a challenging problem commonly addressed through intrinsic rewards. Recent prominent approaches are based on state novelty or variants of artificial curiosity. However, directly applying them to partially observable environments can be ineffective and lead to premature dissipation of intrinsic rewards. Here we propose random curiosity with general value functions (RC-GVF), a novel intrinsic reward function that draws upon connections between these distinct approaches. Instead of using only the current observation's novelty or a curiosity bonus for failing to predict precise environment dynamics, RC-GVF derives intrinsic rewards through predicting temporally extended general value functions. We demonstrate that this improves exploration in a hard-exploration diabolical lock problem. Furthermore, RC-GVF significantly outperforms previous methods in the absence of ground-truth episodic counts in the partially observable MiniGrid environments. Panoramic observations on MiniGrid further boost RC-GVF's performance such that it is competitive to baselines exploiting privileged information in form of episodic counts.