Semantic Labelling and Learning for Parity Game Solving in LTL Synthesis

We propose "semantic labelling" as a novel ingredient for solving games in the context of LTL synthesis. It exploits recent advances in the automata-based approach, yielding more information for each state of the generated parity game than the game graph can capture. We utilize this extra information to improve standard approaches as follows. (i) Compared to strategy improvement (SI) with random initial strategy, a more informed initialization often yields a winning strategy directly without any computation. (ii) This initialization makes SI also yield smaller solutions. (iii) While Q-learning on the game graph turns out not too efficient, Q-learning with the semantic information becomes competitive to SI. Since already the simplest heuristics achieve significant improvements the experimental results demonstrate the utility of semantic labelling. This extra information opens the door to more advanced learning approaches both for initialization and improvement of strategies.

[1]  S. Safra,et al.  On the complexity of omega -automata , 1988, [Proceedings 1988] 29th Annual Symposium on Foundations of Computer Science.

[2]  Alexandre Duret-Lutz,et al.  Spot 2 . 0 — a framework for LTL and ω-automata manipulation , 2016 .

[3]  Bernd Finkbeiner,et al.  The 5th Reactive Synthesis Competition (SYNTCOMP 2018): Benchmarks, Participants & Results , 2019, ArXiv.

[4]  Nir Piterman From Nondeterministic Büchi and Streett Automata to Deterministic Parity Automata , 2007, Log. Methods Comput. Sci..

[5]  Jan Kretínský,et al.  From LTL to Deterministic Automata: A Safraless Compositional Approach , 2014, CAV.

[6]  Marcin Jurdzinski,et al.  A Discrete Strategy Improvement Algorithm for Solving Parity Games , 2000, CAV.

[7]  Amir Pnueli,et al.  Synthesis of Reactive(1) Designs , 2006, VMCAI.

[8]  Jan Kretínský,et al.  Rabinizer 4: From LTL to Your Favourite Deterministic Automaton , 2018, CAV.

[9]  Oliver Friedmann,et al.  Solving Parity Games in Practice , 2009, ATVA.

[10]  Christel Baier,et al.  On-the-Fly Stuttering in the Construction of Deterministic ω-Automata , 2007 .

[11]  Roderick Bloem,et al.  PARTY Parameterized Synthesis of Token Rings , 2013, CAV.

[12]  Orna Kupferman,et al.  Safraless decision procedures , 2005, 46th Annual IEEE Symposium on Foundations of Computer Science (FOCS'05).

[13]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[14]  Maximilien Colange,et al.  Reactive Synthesis from LTL Specification with Spot , 2018 .

[15]  John Fearnley,et al.  Efficient Parallel Strategy Improvement for Parity Games , 2017, CAV.

[16]  Roderick Bloem,et al.  Optimizations for LTL Synthesis , 2006, 2006 Formal Methods in Computer Aided Design.

[17]  C. Baier,et al.  Experiments with Deterministic ω-Automata for Formulas of Linear Temporal Logic , 2005 .

[18]  Sven Schewe,et al.  Solving parity games in big steps , 2007, J. Comput. Syst. Sci..

[19]  Ufuk Topcu,et al.  An Automaton Learning Approach to Solving Safety Games over Infinite Graphs , 2016, TACAS.

[20]  Bernd Finkbeiner,et al.  BoSy: An Experimentation Framework for Bounded Synthesis , 2017, CAV.

[21]  Roderick Bloem,et al.  Anzu: A Tool for Property Synthesis , 2007, CAV.

[22]  Calin Belta,et al.  LTL receding horizon control for finite deterministic systems , 2014, Autom..

[23]  Michael Luttenberger,et al.  Strix: Explicit Reactive Synthesis Strikes Back! , 2018, CAV.

[24]  Orna Kupferman,et al.  Recent Challenges and Ideas in Temporal Synthesis , 2012, SOFSEM.

[25]  Jan Kretínský,et al.  From LTL and Limit-Deterministic Büchi Automata to Deterministic Parity Automata , 2017, TACAS.

[26]  Sven Schewe,et al.  Tighter Bounds for the Determinisation of Büchi Automata , 2009, FoSSaCS.

[27]  Pierre Wolper,et al.  An Automata-Theoretic Approach to Automatic Program Verification (Preliminary Report) , 1986, LICS.

[28]  Tom van Dijk,et al.  Oink: an Implementation and Evaluation of Modern Parity Game Solvers , 2018, TACAS.

[29]  Véronique Bruyère,et al.  Acacia+, a Tool for LTL Synthesis , 2012, CAV.

[30]  Wieslaw Zielonka,et al.  Infinite Games on Finitely Coloured Graphs with Applications to Automata on Infinite Trees , 1998, Theor. Comput. Sci..

[31]  Bernd Finkbeiner,et al.  Bounded synthesis , 2007, International Journal on Software Tools for Technology Transfer.

[32]  Alexandre Duret-Lutz,et al.  Spot 2.0 - A Framework for LTL and \omega -Automata Manipulation , 2016, ATVA.