Shield Synthesis for Reinforcement Learning

Bettina Könighofer, Roderick Bloem, Nils Jansen, Florian Lukas Lorber

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandBegutachtung


Reinforcement learning algorithms discover policies that
maximize reward. However, these policies generally do not adhere to
safety, leaving safety in reinforcement learning (and in artificial intelligence in general) an open research problem. Shield synthesis is a formal
approach to synthesize a correct-by-construction reactive system called a
shield that enforces safety properties of a running system while interfering with its operation as little as possible. A shield attached to a learning
agent guarantees safety during learning and execution phases. In this
paper we summarize three types of shields that are synthesized from
different specification languages, and discuss their applicability to reinforcement learning. First, we discuss deterministic shields that enforce
specifications expressed as linear temporal logic specifications. Second,
we discuss the synthesis of probabilistic shields from specifications in
probabilistic temporal logic. Third, we discuss how to synthesize timed
shields from timed automata specifications. This paper summarizes the
application areas, advantages, disadvantages and synthesis approaches
for the three types of shields and gives an overview of experimental
TitelLeveraging Applications of Formal Methods, Verification and Validation
UntertitelVerification Principles - 9th International Symposium on Leveraging Applications of Formal Methods, ISoLA 2020, Proceedings
Redakteure/-innenTiziana Margaria, Bernhard Steffen
PublikationsstatusElektronische Veröffentlichung vor Drucklegung. - 29 Okt. 2020
Veranstaltung2020 International Symposium on Leveraging Applications of Formal Methods - Virtuell, Griechenland
Dauer: 26 Okt. 202030 Okt. 2020


NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Band12476 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349


Konferenz2020 International Symposium on Leveraging Applications of Formal Methods
KurztitelISoLA 2020

ASJC Scopus subject areas

  • Theoretische Informatik
  • Informatik (insg.)


Untersuchen Sie die Forschungsthemen von „Shield Synthesis for Reinforcement Learning“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren