Model-Based Active Exploration
This addresses the challenge of directed exploration in reinforcement learning, particularly for environments where it is critical, but it appears incremental as it builds on existing model-based and ensemble methods.
The paper tackles the problem of efficient exploration in reinforcement learning by introducing Model-Based Active eXploration (MAX), which uses an ensemble of forward models to plan for observing novel events, resulting in at least an order of magnitude more efficiency than strong baselines in semi-random discrete environments.
Efficient exploration is an unsolved problem in Reinforcement Learning which is usually addressed by reactively rewarding the agent for fortuitously encountering novel situations. This paper introduces an efficient active exploration algorithm, Model-Based Active eXploration (MAX), which uses an ensemble of forward models to plan to observe novel events. This is carried out by optimizing agent behaviour with respect to a measure of novelty derived from the Bayesian perspective of exploration, which is estimated using the disagreement between the futures predicted by the ensemble members. We show empirically that in semi-random discrete environments where directed exploration is critical to make progress, MAX is at least an order of magnitude more efficient than strong baselines. MAX scales to high-dimensional continuous environments where it builds task-agnostic models that can be used for any downstream task.