SafePILCO: a software tool for safe and data-efficient policy synthesis

08/07/2020

∙

SafePILCO is a software tool for safe and data-efficient policy search with reinforcement learning. It extends the known PILCO algorithm, originally written in MATLAB, to support safe learning. We provide a Python implementation and leverage existing libraries that allow the codebase to remain short and modular, which is appropriate for wider use by the verification, reinforcement learning, and control communities.

READ FULL TEXT

SafePILCO: a software tool for safe and data-efficient policy synthesis

Data Generation Method for Learning a Low-dimensional Safe Region in Safe Reinforcement Learning

safe-control-gym: a Unified Benchmark Suite for Safe Learning-based Control and Reinforcement Learning

Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning

Neurosymbolic Reinforcement Learning with Formally Verified Exploration

Data-Efficient Reinforcement Learning in Continuous-State POMDPs

Towards Physically Safe Reinforcement Learning under Supervision

Safe Exploration for Efficient Policy Evaluation and Comparison

SafePILCO: a software tool for safe and data-efficient policy synthesis

Related Research

Data Generation Method for Learning a Low-dimensional Safe Region in Safe Reinforcement Learning

safe-control-gym: a Unified Benchmark Suite for Safe Learning-based Control and Reinforcement Learning

Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning

Neurosymbolic Reinforcement Learning with Formally Verified Exploration

Data-Efficient Reinforcement Learning in Continuous-State POMDPs

Towards Physically Safe Reinforcement Learning under Supervision

Safe Exploration for Efficient Policy Evaluation and Comparison