On the Importance of Exploration for Generalization in Reinforcement Learning
This addresses generalization in RL, a key challenge for deploying agents in diverse real-world settings, though it is incremental by focusing on exploration within existing frameworks.
The paper tackles the problem of generalization in reinforcement learning by emphasizing the role of exploration, showing that it helps agents acquire knowledge useful for unseen environments. It proposes EDE, a value-based method that achieves state-of-the-art results on Procgen and Crafter benchmarks.
Existing approaches for improving generalization in deep reinforcement learning (RL) have mostly focused on representation learning, neglecting RL-specific aspects such as exploration. We hypothesize that the agent's exploration strategy plays a key role in its ability to generalize to new environments. Through a series of experiments in a tabular contextual MDP, we show that exploration is helpful not only for efficiently finding the optimal policy for the training environments but also for acquiring knowledge that helps decision making in unseen environments. Based on these observations, we propose EDE: Exploration via Distributional Ensemble, a method that encourages exploration of states with high epistemic uncertainty through an ensemble of Q-value distributions. Our algorithm is the first value-based approach to achieve state-of-the-art on both Procgen and Crafter, two benchmarks for generalization in RL with high-dimensional observations. The open-sourced implementation can be found at https://github.com/facebookresearch/ede .