Confidence Aware Inverse Constrained Reinforcement Learning
This work addresses the need for confidence-aware constraint estimation in ICRL, which is incremental as it builds on prior methods by adding user-specified confidence levels and data sufficiency checks.
The paper tackles the problem of estimating constraints from expert demonstrations in inverse constraint reinforcement learning (ICRL) by providing a method that outputs constraints with a user-specified confidence level, ensuring they are at least as constraining as the true constraints, and it allows users to determine if more expert data is needed to meet confidence and performance goals.
In coming up with solutions to real-world problems, humans implicitly adhere to constraints that are too numerous and complex to be specified completely. However, reinforcement learning (RL) agents need these constraints to learn the correct optimal policy in these settings. The field of Inverse Constraint Reinforcement Learning (ICRL) deals with this problem and provides algorithms that aim to estimate the constraints from expert demonstrations collected offline. Practitioners prefer to know a measure of confidence in the estimated constraints, before deciding to use these constraints, which allows them to only use the constraints that satisfy a desired level of confidence. However, prior works do not allow users to provide the desired level of confidence for the inferred constraints. This work provides a principled ICRL method that can take a confidence level with a set of expert demonstrations and outputs a constraint that is at least as constraining as the true underlying constraint with the desired level of confidence. Further, unlike previous methods, this method allows a user to know if the number of expert trajectories is insufficient to learn a constraint with a desired level of confidence, and therefore collect more expert trajectories as required to simultaneously learn constraints with the desired level of confidence and a policy that achieves the desired level of performance.