Away from Demonstrations

Specification Inference from Demonstrations

We consider the problem of inferring temporal specifications from demonstrations by an agent interacting with an uncertain, stochastic environment. Such specifications are useful for correct-by-construction control of autonomous systems operating in uncertain environments. Some demonstrations may have errors, and the specification inference method must be robust to them. We provide a novel form...

Learning from Limited Demonstrations

We propose a Learning from Demonstration (LfD) algorithm which leverages expert data, even if they are very few or inaccurate. We achieve this by using both expert data, as well as reinforcement signals gathered through trial-and-error interactions with the environment. The key idea of our approach, Approximate Policy Iteration with Demonstration (APID), is that expert’s suggestions are used to...

Robot Learning from Failed Demonstrations

Robot learning from demonstration (RLfD) seeks to enable lay users to encode desired robot behaviors as autonomous controllers. Current work uses a human’s demonstration of the target task to initialize the robot’s policy, and then improves its performance either through practice (with a known reward function), or additional human interaction. In this article, we focus on the initialization ste...

From the Medicare Hmo Demonstrations

Reinforcement Learning from Imperfect Demonstrations

Robust real-world learning should benefit from both demonstrations and interaction with the environment. Current approaches to learning from demonstration and reward perform supervised learning on expert demonstration data and use reinforcement learning to further improve performance based on reward from the environment. These tasks have divergent losses which are difficult to jointly optimize;...