Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning

Steven Prestwich; Vedat Dogan; Barry O’Sullivan

doi:10.1007/978-3-031-94956-2_4

Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning

Research output: Chapter in Book/Report/Conference proceedings › Conference proceeding › peer-review

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.

Original language	English
Title of host publication	Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings
Editors	Hamid R. Arabnia, Leonidas Deligiannidis, Farzan Shenavarmasouleh, Soheyla Amirian, Farid Ghareh Mohammadi
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	51-63
Number of pages	13
ISBN (Print)	9783031949555
DOIs	https://doi.org/10.1007/978-3-031-94956-2_4
Publication status	Published - 2025
Event	11th International Conference on Computational Science and Computational Intelligence, CSCI 2024 - Las Vegas, United States Duration: 11 Dec 2024 → 13 Dec 2024

Publication series

Name	Communications in Computer and Information Science
Volume	2510 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	11th International Conference on Computational Science and Computational Intelligence, CSCI 2024
Country/Territory	United States
City	Las Vegas
Period	11/12/24 → 13/12/24

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 16 Peace, Justice and Strong Institutions

UCC Futures

Artificial Intelligence and Data Analytics

Keywords

Counterterrorism
Multi-Agent
Multi-Objective
Reinforcement Learning

Access to Document

10.1007/978-3-031-94956-2_4

Cite this

Prestwich, S., Dogan, V., & O’Sullivan, B. (2025). Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning. In H. R. Arabnia, L. Deligiannidis, F. Shenavarmasouleh, S. Amirian, & F. Ghareh Mohammadi (Eds.), Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings (pp. 51-63). (Communications in Computer and Information Science; Vol. 2510 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-94956-2_4

Prestwich, Steven ; Dogan, Vedat ; O’Sullivan, Barry. / Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning. Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings. editor / Hamid R. Arabnia ; Leonidas Deligiannidis ; Farzan Shenavarmasouleh ; Soheyla Amirian ; Farid Ghareh Mohammadi. Springer Science and Business Media Deutschland GmbH, 2025. pp. 51-63 (Communications in Computer and Information Science).

@inproceedings{38c4ef5d60f44ae1bf05787e46daed25,

title = "Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning",

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.",

keywords = "Counterterrorism, Multi-Agent, Multi-Objective, Reinforcement Learning",

author = "Steven Prestwich and Vedat Dogan and Barry O{\textquoteright}Sullivan",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.; 11th International Conference on Computational Science and Computational Intelligence, CSCI 2024 ; Conference date: 11-12-2024 Through 13-12-2024",

year = "2025",

doi = "10.1007/978-3-031-94956-2\_4",

language = "English",

isbn = "9783031949555",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "51--63",

editor = "Arabnia, \{Hamid R.\} and Leonidas Deligiannidis and Farzan Shenavarmasouleh and Soheyla Amirian and \{Ghareh Mohammadi\}, Farid",

booktitle = "Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings",

address = "Germany",

}

Prestwich, S , Dogan, V & O’Sullivan, B 2025, Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning. in HR Arabnia, L Deligiannidis, F Shenavarmasouleh, S Amirian & F Ghareh Mohammadi (eds), Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings. Communications in Computer and Information Science, vol. 2510 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 51-63, 11th International Conference on Computational Science and Computational Intelligence, CSCI 2024, Las Vegas, United States, 11/12/24. https://doi.org/10.1007/978-3-031-94956-2_4

Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning. / Prestwich, Steven ; Dogan, Vedat ; O’Sullivan, Barry.
Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings. ed. / Hamid R. Arabnia; Leonidas Deligiannidis; Farzan Shenavarmasouleh; Soheyla Amirian; Farid Ghareh Mohammadi. Springer Science and Business Media Deutschland GmbH, 2025. p. 51-63 (Communications in Computer and Information Science; Vol. 2510 CCIS).

Research output: Chapter in Book/Report/Conference proceedings › Conference proceeding › peer-review

TY - GEN

T1 - Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning

AU - Prestwich, Steven

AU - Dogan, Vedat

AU - O’Sullivan, Barry

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

PY - 2025

Y1 - 2025

N2 - 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.

AB - 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.

KW - Counterterrorism

KW - Multi-Agent

KW - Multi-Objective

KW - Reinforcement Learning

UR - https://www.scopus.com/pages/publications/105014505956

U2 - 10.1007/978-3-031-94956-2_4

DO - 10.1007/978-3-031-94956-2_4

M3 - Conference proceeding

AN - SCOPUS:105014505956

SN - 9783031949555

T3 - Communications in Computer and Information Science

SP - 51

EP - 63

BT - Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings

A2 - Arabnia, Hamid R.

A2 - Deligiannidis, Leonidas

A2 - Shenavarmasouleh, Farzan

A2 - Amirian, Soheyla

A2 - Ghareh Mohammadi, Farid

PB - Springer Science and Business Media Deutschland GmbH

T2 - 11th International Conference on Computational Science and Computational Intelligence, CSCI 2024

Y2 - 11 December 2024 through 13 December 2024

ER -

Prestwich S , Dogan V , O’Sullivan B. Counterterrorism Planning by Multi-objective Multi-agent Reinforcement Learning. In Arabnia HR, Deligiannidis L, Shenavarmasouleh F, Amirian S, Ghareh Mohammadi F, editors, Computational Science and Computational Intelligence - 11th International Conference, CSCI 2024, Proceedings. Springer Science and Business Media Deutschland GmbH. 2025. p. 51-63. (Communications in Computer and Information Science). doi: 10.1007/978-3-031-94956-2_4