Globally Stable Neural Imitation Policies

arXiv:2403.04118v1 Announce Type: cross
Abstract: Imitation learning presents an effective approach to alleviate the resource-intensive and time-consuming nature of policy learning from scratch in the solution space. Even though the resulting policy can mimic expert demonstrations reliably, it often lacks predictability in unexplored regions of the state-space, giving rise to significant safety concerns in the face of perturbations. To address these challenges, we introduce the Stable Neural Dynamical System (SNDS), an imitation learning regime which produces a policy with formal …

abstract arxiv concerns cs.lg cs.ro expert giving imitation learning nature policy safety scratch solution space state type