In this paper, we propose SAMBA, a novel framework for safe reinforcement learning that combines aspects from probabilistic modelling, information theory, and statistics. Our method builds upon PILCO to enable active exploration using novel(semi-)metrics for out-of-sample Gaussian process evaluation optimised through a multi-objective problem that supports conditional-value-at-risk constraints. We evaluate our algorithm on a variety of safe dynamical system benchmarks involving both low and high-dimensional state representations. Our results show orders of magnitude reductions in samples and violations compared to state-of-the-art methods. Lastly, we provide intuition as to the effectiveness of the framework by a detailed analysis of our active metrics and safety constraints.
@article{arxiv.2006.09436,
title = {SAMBA: Safe Model-Based & Active Reinforcement Learning},
author = {Alexander I. Cowen-Rivers and Daniel Palenicek and Vincent Moens and Mohammed Abdullah and Aivar Sootla and Jun Wang and Haitham Ammar},
journal= {arXiv preprint arXiv:2006.09436},
year = {2020}
}