All Models are Wrong, Knowing Where is Useful: On Model Uncertainty in Reinforcement Learning

Property details

Additional property details will be updated shortly.

Property description

Model-based reinforcement learning (MBRL) infers information about the environment from a learned dynamics model and bears the potential to address open problems such as data efficient and safe learning in robotics. However, inaccuracies of the learned dynamics model are typically exploited by the agent, substantially hampering the capabilities of MBRL methods. We present a framework for dealing with inaccuracies of probabilistic models through targeted handling of uncertainty that effectively mitigates model exploitation. We present recent successes in learning directly on hardware and safe exploration, and discuss future directions for uncertainty-aware MBRL.

Links and sources

• Official / arXiv page
• PDF from original source

Comments