Counterterrorism planning by multi-objective multi-agent reinforcement learning

Thumbnail Image
Files
CSCI.pdf(338.6 KB)
Accepted Version
Date
2025
Authors
Prestwich, Steven
Dogan, Vedat
O'Sullivan, Barry
Journal Title
Journal ISSN
Volume Title
Publisher
Springer Nature
Published Version
10.1007/978-3-031-94956-2_4
Research Projects
Organizational Units
Journal Issue
Abstract
In areas including counterterrorism, security, diplomacy and supply chain optimisation, an analyst must make decisions under assumptions about the risks posed by an adversary. Research fields including operations research, decision theory, game theory, influence diagrams and adversarial risk analysis provide a rich variety of methods to model and solve such problems. Reinforcement learning (RL) is also an approach to sequential decision making that has been applied to specific problems involving risk. We propose multi-objective multi-agent RL (MOMARL) as a general-purpose approach to risk analysis. Using a MOMARL solver we model and solve variants of a problem in counterterrorism, including a notoriously difficult problem class: pessimistic bilevel optimisation under uncertainty.
Description
Keywords
Counterterrorism , Reinforcement learning , Multi-objective , Multi-agent
Citation
Prestwich, S., Dogan, V. and O'Sullivan, B. (2025) 'Counterterrorism planning by multi-objective multi-agent reinforcement learning', in Arabnia, H. R., Deligiannidis, L., Shenavarmasouleh, F., Amirian, S., Ghareh Mohammadi, F. (eds) Computational Science and Computational Intelligence. CSCI 2024. Communications in Computer and Information Science, 2510, pp. 51-63. Cham: Springer. https://doi.org/10.1007/978-3-031-94956-2_4
Link to publisher’s version
URI
https://hdl.handle.net/10468/17864
Collections
Insight Centre for Data Analytics - Conference Items
Copyright
© 2025, The Authors, under exclusive license to Springer Nature Switzerland AG. This version of the paper has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/978-3-031-94956-2_4