Towards Interpretable Reinforcement Learning with Constrained Normalizing Flow Policies

arXiv:2405.01198v1 Announce Type: new
Abstract: Reinforcement learning policies are typically represented by black-box neural networks, which are non-interpretable and not well-suited for safety-critical domains. To address both of these issues, we propose constrained normalizing flow policies as interpretable and safe-by-construction policy models. We achieve safety for reinforcement learning problems with instantaneous safety constraints, for which we can exploit domain knowledge by analytically constructing a normalizing flow that ensures constraint satisfaction. The normalizing flow corresponds to an interpretable sequence of transformations …

abstract arxiv box construction cs.ai cs.lg domains flow networks neural networks policies policy reinforcement reinforcement learning safe safety safety-critical type