Online shielding for reinforcement learning
Open Access
- 23 September 2022
- journal article
- research article
- Published by Springer Nature in Innovations in Systems and Software Engineering
- Vol. 19 (4) , 379-394
Abstract
Besides the recent impressive results on reinforcement learning (RL), safety is still one of the major research challenges in RL. RL is a machine-learning approach to determine near-optimal policies in Markov decision processes (MDPs). In this paper, we consider the setting where the safety-relevant fragment of the MDP together with a temporal logic safety specification is given, and many safety violations can be avoided by planning ahead a short time into the future. We propose an approach for online safety shielding of RL agents. During runtime, the shield analyses the safety of each available action. For any action, the shield computes the maximal probability to not violate the safety specification within the next k steps when executing this action. Based on this probability and a given threshold, the shield decides whether to block an action from the agent. Existing offline shielding approaches compute exhaustively the safety of all state-action combinations ahead of time, resulting in huge computation times and large memory consumption. The intuition behind online shielding is to compute at runtime the set of all states that could be reached in the near future. For each of these states, the safety of all available actions is analysed and used for shielding as soon as one of the considered states is reached. Our approach is well-suited for high-level planning problems where the time between decisions can be used for safety computations and it is sustainable for the agent to wait until these computations are finished. For our evaluation, we selected a 2-player version of the classical computer game Snake. The game represents a high-level planning problem that requires fast decisions and the multiplayer setting induces a large state space, which is computationally expensive to analyse exhaustively.Keywords
All Related Versions
Funding Information
- Horizon 2020 (956123)
- Silicon Austria Labs
- Technische Universität Graz
- Lamarr Security Research
This publication has 27 references indexed in Scilit:
- Verifiably Safe Off-Model Reinforcement Learning2019
- Planning for cars that coordinate with people: leveraging effects on human actions for planning and active information gathering over human internal stateAutonomous Robots, 2018
- A Storm is Coming: A Modern Probabilistic Model Checker2017
- The Probabilistic Model Checking Landscape2016
- Learning deterministic probabilistic automata from a model checking perspectiveMachine Learning, 2016
- Mastering the game of Go with deep neural networks and tree searchNature, 2016
- Shield Synthesis:2015
- PRISM 4.0: Verification of Probabilistic Real-Time Systems2011
- Model checking for probability and time: from theory to practice2003
- The temporal logic of programs1977