Regstar: efficient strategy synthesis for adversarial patrolling games

David Klaška, Antonín Kučera, Vít Musil, Vojtěch Řehák
Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, PMLR 161:471-481, 2021.

Abstract

We design a new efficient strategy synthesis method applicable to adversarial patrolling problems on graphs with arbitrary-length edges and possibly imperfect intrusion detection. The core ingredient is an efficient algorithm for computing the value and the gradient of a function assigning to every strategy its “protection” achieved. This allows for designing an efficient strategy improvement algorithm by differentiable programming and optimization techniques. Our method is the first one applicable to real-world patrolling graphs of reasonable sizes. It outperforms the state-of-the-art strategy synthesis algorithm by a margin.

Cite this Paper

BibTeX
@InProceedings{pmlr-v161-klaska21a, title = {Regstar: efficient strategy synthesis for adversarial patrolling games}, author = {Kla\v{s}ka, David and Ku\v{c}era, Anton\'{i}n and Musil, V\'{i}t and \v{R}eh\'{a}k, Vojt\v{e}ch}, booktitle = {Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence}, pages = {471--481}, year = {2021}, editor = {de Campos, Cassio and Maathuis, Marloes H.}, volume = {161}, series = {Proceedings of Machine Learning Research}, month = {27--30 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v161/klaska21a/klaska21a.pdf}, url = {https://proceedings.mlr.press/v161/klaska21a.html}, abstract = {We design a new efficient strategy synthesis method applicable to adversarial patrolling problems on graphs with arbitrary-length edges and possibly imperfect intrusion detection. The core ingredient is an efficient algorithm for computing the value and the gradient of a function assigning to every strategy its “protection” achieved. This allows for designing an efficient strategy improvement algorithm by differentiable programming and optimization techniques. Our method is the first one applicable to real-world patrolling graphs of reasonable sizes. It outperforms the state-of-the-art strategy synthesis algorithm by a margin.} }
Endnote
%0 Conference Paper %T Regstar: efficient strategy synthesis for adversarial patrolling games %A David Klaška %A Antonín Kučera %A Vít Musil %A Vojtěch Řehák %B Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2021 %E Cassio de Campos %E Marloes H. Maathuis %F pmlr-v161-klaska21a %I PMLR %P 471--481 %U https://proceedings.mlr.press/v161/klaska21a.html %V 161 %X We design a new efficient strategy synthesis method applicable to adversarial patrolling problems on graphs with arbitrary-length edges and possibly imperfect intrusion detection. The core ingredient is an efficient algorithm for computing the value and the gradient of a function assigning to every strategy its “protection” achieved. This allows for designing an efficient strategy improvement algorithm by differentiable programming and optimization techniques. Our method is the first one applicable to real-world patrolling graphs of reasonable sizes. It outperforms the state-of-the-art strategy synthesis algorithm by a margin.
APA
Klaška, D., Kučera, A., Musil, V. & Řehák, V.. (2021). Regstar: efficient strategy synthesis for adversarial patrolling games. Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 161:471-481 Available from https://proceedings.mlr.press/v161/klaska21a.html.

Related Material