The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation

被引：22

作者：

Lucas, Simon M. ^{[1
]}

Liu, Jialin ^{[1
]}

Perez-Liebana, Diego ^{[1
]}

机构：

[1] Queen Mary Univ London, Sch Elect Engn & Comp Sci, London, England

来源：

2018 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC) | 2018年

关键词：

Estimation of Distribution Algorithm; Evolutionary Algorithm; Hyper-Parameter Optimisation; Rolling Horizon Evolution; Game Playing Agent; Noisy Optimisation;

D O I：

10.1109/CEC.2018.8477869

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper describes the N-Tuple Bandit Evolutionary Algorithm (NTBEA), an optimisation algorithm developed for noisy and expensive discrete (combinatorial) optimisation problems. The algorithm is applied to two game-based hyperparameter optimisation problems. The N-Tuple system directly models the statistics, approximating the fitness and number of evaluations of each modelled combination of parameters. The model is simple, efficient and informative. Results show that the NTBEA significantly outperforms grid search and an estimation of distribution algorithm.

引用

页码：221 / 229

页数：9

共 34 条

[31] Perez-Liebana D, 2016, AAAI CONF ARTIF INTE, P4335
[32] A theoretical and experimental account of n-tuple classifier performance
Rohwer, R
Morciniec, M
[J]. NEURAL COMPUTATION, 1996, 8 (03) : 629 - 642
[33] Bandit-Based Estimation of Distribution Algorithms for Noisy Optimization: Rigorous Runtime Analysis
Rolet, Philippe
Teytaud, Olivier
[J]. LEARNING AND INTELLIGENT OPTIMIZATION, 2010, 6073 : 97 - 110
[34] Preference Learning for Move Prediction and Evaluation Function Approximation in Othello
Runarsson, Thomas Philip
Lucas, Simon M.
[J]. IEEE TRANSACTIONS ON COMPUTATIONAL INTELLIGENCE AND AI IN GAMES, 2014, 6 (03) : 300 - 313

← 1 2 3 4 →