TY - GEN
T1 - A Comprehensive Framework for Turn-Taking Evaluation in Multi-agent Systems
T2 - 24th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2025
AU - Papadopoulos, Nikolaos Al
AU - Taratori, Rallou
AU - Sanchez-Fibla, Marti
AU - Psannis, Konstantinos E.
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2025/11/1
Y1 - 2025/11/1
N2 - Turn-taking is a cornerstone of coordination in multi-agent systems, ensuring fairness, efficiency, and conflict avoidance. While detailed metrics like ALT measures offer precise turn-taking analysis, their computational complexity limits scalability. We introduce Rotational Periodicity (RP), a lightweight metric combining Average Waiting Episodes (AWE) and Waiting Periods Evaluation (WPE), and propose a comprehensive framework for evaluating turn-taking against ideal patterns such as Perfect Alternation (PA). Evaluated in scenarios like the multi-agent Battle of the Exes (MBoE), RP integrates with Reward Fairness (RF) and Efficiency (E) to form a versatile framework for assessing and inducing coordination. This paper outlines this toolkit, emphasizing its adaptability for various turn-taking contexts and its computational efficiency for large-scale multi-agent simulations.
AB - Turn-taking is a cornerstone of coordination in multi-agent systems, ensuring fairness, efficiency, and conflict avoidance. While detailed metrics like ALT measures offer precise turn-taking analysis, their computational complexity limits scalability. We introduce Rotational Periodicity (RP), a lightweight metric combining Average Waiting Episodes (AWE) and Waiting Periods Evaluation (WPE), and propose a comprehensive framework for evaluating turn-taking against ideal patterns such as Perfect Alternation (PA). Evaluated in scenarios like the multi-agent Battle of the Exes (MBoE), RP integrates with Reward Fairness (RF) and Efficiency (E) to form a versatile framework for assessing and inducing coordination. This paper outlines this toolkit, emphasizing its adaptability for various turn-taking contexts and its computational efficiency for large-scale multi-agent simulations.
KW - Aggregation Functions
KW - Decision Making
KW - Efficiency
KW - Fairness
KW - Multi-Agent Systems
KW - Rotational Periodicity
KW - Turn-Taking
UR - https://www.scopus.com/pages/publications/105022173035
U2 - 10.1007/978-3-032-03711-4_16
DO - 10.1007/978-3-032-03711-4_16
M3 - Conference contribution
AN - SCOPUS:105022173035
SN - 9783032037107
T3 - Lecture Notes in Computer Science
SP - 191
EP - 203
BT - Artificial Intelligence and Soft Computing - 24th International Conference, ICAISC 2025, Proceedings
A2 - Rutkowski, Leszek
A2 - Scherer, Rafal
A2 - Korytkowski, Marcin
A2 - Pedrycz, Witold
A2 - Tadeusiewicz, Ryszard
A2 - Zurada, Jacek M.
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 22 June 2025 through 26 June 2025
ER -