Making Curiosity Explicit in Vision-based RL
This work addresses sample efficiency and generalization issues in vision-based RL for control tasks, representing an incremental improvement over existing state representation learning methods.
The paper tackles the problem of low sample efficiency in vision-based reinforcement learning by improving sample diversity through enhanced exploration, resulting in outperforming baselines across all tested environments with stabilized training, reduced reward variance, and boosted efficiency.
Vision-based reinforcement learning (RL) is a promising technique to solve control tasks involving images as the main observation. State-of-the-art RL algorithms still struggle in terms of sample efficiency, especially when using image observations. This has led to an increased attention on integrating state representation learning (SRL) techniques into the RL pipeline. Work in this field demonstrates a substantial improvement in sample efficiency among other benefits. However, to take full advantage of this paradigm, the quality of samples used for training plays a crucial role. More importantly, the diversity of these samples could affect the sample efficiency of vision-based RL, but also its generalization capability. In this work, we present an approach to improve the sample diversity. Our method enhances the exploration capability of the RL algorithms by taking advantage of the SRL setup. Our experiments show that the presented approach outperforms the baseline for all tested environments. These results are most apparent for environments where the baseline method struggles. Even in simple environments, our method stabilizes the training, reduces the reward variance and boosts sample efficiency.