Timezone: »

Learning Safe Policies with Expert Guidance
Jessie Huang · Fa Wu · Doina Precup · Yang Cai

Wed Dec 05 07:45 AM -- 09:45 AM (PST) @ Room 517 AB #163

We propose a framework for ensuring safe behavior of a reinforcement learning agent when the reward function may be difficult to specify. In order to do this, we rely on the existence of demonstrations from expert policies, and we provide a theoretical framework for the agent to optimize in the space of rewards consistent with its existing knowledge. We propose two methods to solve the resulting optimization: an exact ellipsoid-based method and a method in the spirit of the "follow-the-perturbed-leader" algorithm. Our experiments demonstrate the behavior of our algorithm in both discrete and continuous problems. The trained agent safely avoids states with potential negative effects while imitating the behavior of the expert in the other states.

Author Information

Jessie Huang (McGill University)
Fa Wu (McGill)
Doina Precup (McGill University / DeepMind Montreal)
Yang Cai (Yale University)

More from the Same Authors