A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem

Pengpeng Wang, Yifeng Zeng, Bilian Chen, Langcai Cao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

14 Downloads (Pure)

Abstract

Production constrained build-order optimization problems challenge artificial intelligence research in computer game applications due to an uncertain set of constraints. Traditional approaches provide subjective values in the constraint formulation therefore resulting in unexpected performance of the optimal build-order in a game. In this article, we propose a data-driven approach to solve a build-order optimization problem in StarCraft. We formulate the constraint by learning the parameter values from game replay data, which complements more precise problem formulation. To solve the optimization, we use the improved genetic algorithm by learning initial solutions from the data. We show the performance of the data-driven methods in a StarCraft simulation platform.
Original languageEnglish
Title of host publicationProceedings of the 38th Chinese Control Conference
PublisherIEEE
Pages2692-2697
ISBN (Electronic)9789881563972
DOIs
Publication statusPublished - 17 Oct 2019
Event2019 Chinese Control Conference - Guangzhou, China
Duration: 27 Jul 201930 Jul 2019

Publication series

NameChinese Control Conference (CCC)
PublisherIEEE
Volume2019
ISSN (Electronic)1934-1768

Conference

Conference2019 Chinese Control Conference
Abbreviated titleCCC
CountryChina
CityGuangzhou
Period27/07/1930/07/19

Fingerprint

Computer games
Artificial intelligence
Genetic algorithms

Cite this

Wang, P., Zeng, Y., Chen, B., & Cao, L. (2019). A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem. In Proceedings of the 38th Chinese Control Conference (pp. 2692-2697). (Chinese Control Conference (CCC); Vol. 2019). IEEE. https://doi.org/10.23919/ChiCC.2019.8866045
Wang, Pengpeng ; Zeng, Yifeng ; Chen, Bilian ; Cao, Langcai. / A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem. Proceedings of the 38th Chinese Control Conference. IEEE, 2019. pp. 2692-2697 (Chinese Control Conference (CCC)).
@inproceedings{5fd32f07833a495b9e76ed0a62757ee7,
title = "A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem",
abstract = "Production constrained build-order optimization problems challenge artificial intelligence research in computer game applications due to an uncertain set of constraints. Traditional approaches provide subjective values in the constraint formulation therefore resulting in unexpected performance of the optimal build-order in a game. In this article, we propose a data-driven approach to solve a build-order optimization problem in StarCraft. We formulate the constraint by learning the parameter values from game replay data, which complements more precise problem formulation. To solve the optimization, we use the improved genetic algorithm by learning initial solutions from the data. We show the performance of the data-driven methods in a StarCraft simulation platform.",
author = "Pengpeng Wang and Yifeng Zeng and Bilian Chen and Langcai Cao",
year = "2019",
month = "10",
day = "17",
doi = "10.23919/ChiCC.2019.8866045",
language = "English",
series = "Chinese Control Conference (CCC)",
publisher = "IEEE",
pages = "2692--2697",
booktitle = "Proceedings of the 38th Chinese Control Conference",

}

Wang, P, Zeng, Y, Chen, B & Cao, L 2019, A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem. in Proceedings of the 38th Chinese Control Conference. Chinese Control Conference (CCC), vol. 2019, IEEE, pp. 2692-2697, 2019 Chinese Control Conference , Guangzhou, China, 27/07/19. https://doi.org/10.23919/ChiCC.2019.8866045

A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem. / Wang, Pengpeng; Zeng, Yifeng; Chen, Bilian; Cao, Langcai.

Proceedings of the 38th Chinese Control Conference. IEEE, 2019. p. 2692-2697 (Chinese Control Conference (CCC); Vol. 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem

AU - Wang, Pengpeng

AU - Zeng, Yifeng

AU - Chen, Bilian

AU - Cao, Langcai

PY - 2019/10/17

Y1 - 2019/10/17

N2 - Production constrained build-order optimization problems challenge artificial intelligence research in computer game applications due to an uncertain set of constraints. Traditional approaches provide subjective values in the constraint formulation therefore resulting in unexpected performance of the optimal build-order in a game. In this article, we propose a data-driven approach to solve a build-order optimization problem in StarCraft. We formulate the constraint by learning the parameter values from game replay data, which complements more precise problem formulation. To solve the optimization, we use the improved genetic algorithm by learning initial solutions from the data. We show the performance of the data-driven methods in a StarCraft simulation platform.

AB - Production constrained build-order optimization problems challenge artificial intelligence research in computer game applications due to an uncertain set of constraints. Traditional approaches provide subjective values in the constraint formulation therefore resulting in unexpected performance of the optimal build-order in a game. In this article, we propose a data-driven approach to solve a build-order optimization problem in StarCraft. We formulate the constraint by learning the parameter values from game replay data, which complements more precise problem formulation. To solve the optimization, we use the improved genetic algorithm by learning initial solutions from the data. We show the performance of the data-driven methods in a StarCraft simulation platform.

U2 - 10.23919/ChiCC.2019.8866045

DO - 10.23919/ChiCC.2019.8866045

M3 - Conference contribution

T3 - Chinese Control Conference (CCC)

SP - 2692

EP - 2697

BT - Proceedings of the 38th Chinese Control Conference

PB - IEEE

ER -

Wang P, Zeng Y, Chen B, Cao L. A Data-driven Approach to Solve a Production Constrained Build-order Optimization Problem. In Proceedings of the 38th Chinese Control Conference. IEEE. 2019. p. 2692-2697. (Chinese Control Conference (CCC)). https://doi.org/10.23919/ChiCC.2019.8866045