Active Learning of Dynamics Using Prior Domain Knowledge in the Sampling Process
This work addresses the challenge of efficiently learning dynamics in control systems, though it appears incremental by building on existing active learning methods.
The paper tackles the problem of learning dynamics by incorporating prior domain knowledge into the active sampling process, resulting in accelerated learning and reduced model uncertainty, as demonstrated on an under-actuated pendulum and half-cheetah MuJoCo environment.
We present an active learning algorithm for learning dynamics that leverages side information by explicitly incorporating prior domain knowledge into the sampling process. Our proposed algorithm guides the exploration toward regions that demonstrate high empirical discrepancy between the observed data and an imperfect prior model of the dynamics derived from side information. Through numerical experiments, we demonstrate that this strategy explores regions of high discrepancy and accelerates learning while simultaneously reducing model uncertainty. We rigorously prove that our active learning algorithm yields a consistent estimate of the underlying dynamics by providing an explicit rate of convergence for the maximum predictive variance. We demonstrate the efficacy of our approach on an under-actuated pendulum system and on the half-cheetah MuJoCo environment.