Standard
Harvard
Pelcner, L, do Carmo Alves, MA, Soriano Marcolino, L, Harrison, P
& Atkinson, P 2024,
Incentive-Driven Multi-agent Reinforcement Learning Approach for Commons Dilemmas in Land-Use. in R Ariska, V Sanchez-Anguix, S Stein, R Aydoğan, L van der Torre & T Ito (eds),
PRIMA 2024: Principles and Practice of Multi-Agent Systems: : 25th International Conference, Kyoto, Japan, November 18–24, 2024, Proceedings. Lecture Notes in Computer Science, vol. 15395, Springer, Cham, pp. 284-289.
https://doi.org/10.1007/978-3-031-77367-9_21
APA
Pelcner, L., do Carmo Alves, M. A., Soriano Marcolino, L., Harrison, P.
, & Atkinson, P. (2024).
Incentive-Driven Multi-agent Reinforcement Learning Approach for Commons Dilemmas in Land-Use. In R. Ariska, V. Sanchez-Anguix, S. Stein, R. Aydoğan, L. van der Torre, & T. Ito (Eds.),
PRIMA 2024: Principles and Practice of Multi-Agent Systems: : 25th International Conference, Kyoto, Japan, November 18–24, 2024, Proceedings (pp. 284-289). (Lecture Notes in Computer Science; Vol. 15395). Springer.
https://doi.org/10.1007/978-3-031-77367-9_21
Vancouver
Pelcner L, do Carmo Alves MA, Soriano Marcolino L, Harrison P
, Atkinson P.
Incentive-Driven Multi-agent Reinforcement Learning Approach for Commons Dilemmas in Land-Use. In Ariska R, Sanchez-Anguix V, Stein S, Aydoğan R, van der Torre L, Ito T, editors, PRIMA 2024: Principles and Practice of Multi-Agent Systems: : 25th International Conference, Kyoto, Japan, November 18–24, 2024, Proceedings. Cham: Springer. 2024. p. 284-289. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-77367-9_21
Author
Bibtex
@inproceedings{01785cb2c1d34f518b82ea35ebcbbe53,
title = "Incentive-Driven Multi-agent Reinforcement Learning Approach for Commons Dilemmas in Land-Use",
abstract = "We propose ORAA, a novel incentive-driven algorithm that guides agents in a property-based Multi-Agent Reinforcement Learning domain to act sustainably considering a common pool of resources in an online manner. ORAA implements our proposed P-MADDPG model to learn and make decisions over the decentralised agents. We test our solutions in our novel domain, the “Pollinators{\textquoteright} Game”, which simulates a property-based scenario and the incentivisation dynamics. We show significant improvement in the incentives{\textquoteright} cost-efficiency, reducing the budget spent while increasing the collection of rewards by individual agents. Besides that, our application shows better results when using learned (approximated) models instead of using and simulating the true models of each agent for planning, saving up to 50% of the available budget for incentivisation.",
author = "Lukasz Pelcner and {do Carmo Alves}, {Matheus Aparecido} and {Soriano Marcolino}, Leandro and Paula Harrison and Peter Atkinson",
year = "2024",
month = nov,
day = "16",
doi = "10.1007/978-3-031-77367-9_21",
language = "English",
isbn = "9783031773662",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "284--289",
editor = "Ryuta Ariska and Victor Sanchez-Anguix and Sebastian Stein and Reyhan Aydoğan and {van der Torre}, Leon and Takayuki Ito",
booktitle = "PRIMA 2024: Principles and Practice of Multi-Agent Systems:",
}
RIS
TY - GEN
T1 - Incentive-Driven Multi-agent Reinforcement Learning Approach for Commons Dilemmas in Land-Use
AU - Pelcner, Lukasz
AU - do Carmo Alves, Matheus Aparecido
AU - Soriano Marcolino, Leandro
AU - Harrison, Paula
AU - Atkinson, Peter
PY - 2024/11/16
Y1 - 2024/11/16
N2 - We propose ORAA, a novel incentive-driven algorithm that guides agents in a property-based Multi-Agent Reinforcement Learning domain to act sustainably considering a common pool of resources in an online manner. ORAA implements our proposed P-MADDPG model to learn and make decisions over the decentralised agents. We test our solutions in our novel domain, the “Pollinators’ Game”, which simulates a property-based scenario and the incentivisation dynamics. We show significant improvement in the incentives’ cost-efficiency, reducing the budget spent while increasing the collection of rewards by individual agents. Besides that, our application shows better results when using learned (approximated) models instead of using and simulating the true models of each agent for planning, saving up to 50% of the available budget for incentivisation.
AB - We propose ORAA, a novel incentive-driven algorithm that guides agents in a property-based Multi-Agent Reinforcement Learning domain to act sustainably considering a common pool of resources in an online manner. ORAA implements our proposed P-MADDPG model to learn and make decisions over the decentralised agents. We test our solutions in our novel domain, the “Pollinators’ Game”, which simulates a property-based scenario and the incentivisation dynamics. We show significant improvement in the incentives’ cost-efficiency, reducing the budget spent while increasing the collection of rewards by individual agents. Besides that, our application shows better results when using learned (approximated) models instead of using and simulating the true models of each agent for planning, saving up to 50% of the available budget for incentivisation.
U2 - 10.1007/978-3-031-77367-9_21
DO - 10.1007/978-3-031-77367-9_21
M3 - Conference contribution/Paper
SN - 9783031773662
T3 - Lecture Notes in Computer Science
SP - 284
EP - 289
BT - PRIMA 2024: Principles and Practice of Multi-Agent Systems:
A2 - Ariska, Ryuta
A2 - Sanchez-Anguix, Victor
A2 - Stein, Sebastian
A2 - Aydoğan, Reyhan
A2 - van der Torre, Leon
A2 - Ito, Takayuki
PB - Springer
CY - Cham
ER -
