Attention-based Curiosity-driven Exploration in Deep Reinforcement Learning
This work addresses exploration challenges in reinforcement learning for agents in sparse-reward environments, presenting incremental improvements over existing curiosity methods.
The paper tackles sparse extrinsic feedback in reinforcement learning by proposing attention-based curiosity-driven exploration methods, including AttA2C and a modified ICM with rational curiosity, achieving improved exploration in experiments.
Reinforcement Learning enables to train an agent via interaction with the environment. However, in the majority of real-world scenarios, the extrinsic feedback is sparse or not sufficient, thus intrinsic reward formulations are needed to successfully train the agent. This work investigates and extends the paradigm of curiosity-driven exploration. First, a probabilistic approach is taken to exploit the advantages of the attention mechanism, which is successfully applied in other domains of Deep Learning. Combining them, we propose new methods, such as AttA2C, an extension of the Actor-Critic framework. Second, another curiosity-based approach - ICM - is extended. The proposed model utilizes attention to emphasize features for the dynamic models within ICM, moreover, we also modify the loss function, resulting in a new curiosity formulation, which we call rational curiosity. The corresponding implementation can be found at https://github.com/rpatrik96/AttA2C/.