Maximum Likelihood Constraint Inference from Stochastic Demonstrations
This work addresses the challenge of extracting implicit constraint information from expert demonstrations in uncertain and risky real-world systems, such as perilous dynamic systems, by generalizing prior deterministic methods to stochastic models.
The paper tackles the problem of inferring constraints from expert demonstrations in stochastic dynamic systems by extending maximum likelihood constraint inference to stochastic applications using maximum causal entropy likelihoods, and proposes an efficient algorithm that computes constraint likelihood and risk tolerance in a unified Bellman backup without increasing computational complexity.
When an expert operates a perilous dynamic system, ideal constraint information is tacitly contained in their demonstrated trajectories and controls. The likelihood of these demonstrations can be computed, given the system dynamics and task objective, and the maximum likelihood constraints can be identified. Prior constraint inference work has focused mainly on deterministic models. Stochastic models, however, can capture the uncertainty and risk tolerance that are often present in real systems of interest. This paper extends maximum likelihood constraint inference to stochastic applications by using maximum causal entropy likelihoods. Furthermore, we propose an efficient algorithm that computes constraint likelihood and risk tolerance in a unified Bellman backup, allowing us to generalize to stochastic systems without increasing computational complexity.