Exploiting Model Errors for Exploration in Model-Based Reinforcement Learning

We address the problem of exploration in model-based reinforcement learning (MBRL). We present Model-Corrective eXploration (MCX) a novel approach to exploration in MBRL that is both agnostic to the model representation and scalable to complex environments. MCX learns to generalise model prediction errors in order to make hypotheses about how the model might else be wrong, and uses such hypotheses for performing planning to facilitate exploration. We demonstrate the efficacy of our method in visual control tasks with the state-of-the-art MBRL algorithm, DreamerV3.

Authors

• Jared Swift
• Matteo Leonetti King's College London

Keywords

• Exploration
• Model-Based Reinforcement Learning
• Reinforcement Learning