TY - GEN
T1 - Portfolio Search and Optimization for General Strategy Game-Playing
AU - Dockhorn, Alexander
AU - Hurtado-Grueso, Jorge
AU - Jeurissen, Dominik
AU - Xu, Linjie
AU - Perez-Liebana, Diego
N1 - Funding Information:
ACKNOWLEDGEMENTS This work is supported by UK EPSRC research grant EP/T008962/1 (https://gaigresearch.github.io/afm/).
PY - 2021
Y1 - 2021
N2 - Portfolio methods represent a simple but efficient type of action abstraction which has shown to improve the performance of search-based agents in a range of strategy games. We first review existing portfolio techniques and propose a new algorithm for optimization and action-selection based on the Rolling Horizon Evolutionary Algorithm. Moreover, a series of variants are developed to solve problems in different aspects. We further analyze the performance of discussed agents in a general strategy game-playing task. For this purpose, we run experiments on three different game-modes of the STRATEGA framework. For the optimization of the agents' parameters and portfolio sets we study the use of the N-tuple Bandit Evolutionary Algorithm. The resulting portfolio sets suggest a high diversity in play-styles while being able to consistently beat the sample agents. An analysis of the agents' performance shows that the proposed algorithm generalizes well to all game-modes and is able to outperform other portfolio methods.
AB - Portfolio methods represent a simple but efficient type of action abstraction which has shown to improve the performance of search-based agents in a range of strategy games. We first review existing portfolio techniques and propose a new algorithm for optimization and action-selection based on the Rolling Horizon Evolutionary Algorithm. Moreover, a series of variants are developed to solve problems in different aspects. We further analyze the performance of discussed agents in a general strategy game-playing task. For this purpose, we run experiments on three different game-modes of the STRATEGA framework. For the optimization of the agents' parameters and portfolio sets we study the use of the N-tuple Bandit Evolutionary Algorithm. The resulting portfolio sets suggest a high diversity in play-styles while being able to consistently beat the sample agents. An analysis of the agents' performance shows that the proposed algorithm generalizes well to all game-modes and is able to outperform other portfolio methods.
KW - General strategy game-playing
KW - N-tuple bandit evolutionary algorithm
KW - Portfolio methods
KW - Stratega
UR - https://www.scopus.com/pages/publications/85110943253
U2 - 10.1109/CEC45853.2021.9504824
DO - 10.1109/CEC45853.2021.9504824
M3 - Conference contribution
AN - SCOPUS:85110943253
T3 - 2021 IEEE Congress on Evolutionary Computation, CEC 2021 - Proceedings
SP - 2085
EP - 2092
BT - 2021 IEEE Congress on Evolutionary Computation, CEC 2021 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 IEEE Congress on Evolutionary Computation, CEC 2021
Y2 - 28 June 2021 through 1 July 2021
ER -