Deep learning for video game playing

Niels Justesen; Philip Bontrager; Julian Togelius; Sebastian Risi

doi:10.1109/TG.2019.2896986

Deep learning for video game playing

Niels Justesen, Philip Bontrager, Julian Togelius, Sebastian Risi

Research output: Contribution to journal › Article › peer-review

Abstract

In this paper, we review recent deep learning advances in the context of how they have been applied to play different types of video games such as first-person shooters, arcade games, and real-time strategy games. We analyze the unique requirements that different game genres pose to a deep learning system and highlight important open challenges in the context of applying these machine learning methods to video games, such as general game playing, dealing with extremely large decision spaces and sparse rewards.

Original language	English (US)
Article number	2896986
Pages (from-to)	1-20
Number of pages	20
Journal	IEEE Transactions on Games
Volume	12
Issue number	1
DOIs	https://doi.org/10.1109/TG.2019.2896986
State	Published - Mar 2020

Keywords

Algorithms
Artificial intelligence
Learning
Machine learning algorithms
Multilayer neural network

ASJC Scopus subject areas

Software
Control and Systems Engineering
Artificial Intelligence
Electrical and Electronic Engineering

Access to Document

10.1109/TG.2019.2896986

Cite this

@article{9b01f9458e2248b9955f32c3e6f3d926,

title = "Deep learning for video game playing",

abstract = "In this paper, we review recent deep learning advances in the context of how they have been applied to play different types of video games such as first-person shooters, arcade games, and real-time strategy games. We analyze the unique requirements that different game genres pose to a deep learning system and highlight important open challenges in the context of applying these machine learning methods to video games, such as general game playing, dealing with extremely large decision spaces and sparse rewards.",

keywords = "Algorithms, Artificial intelligence, Learning, Machine learning algorithms, Multilayer neural network",

author = "Niels Justesen and Philip Bontrager and Julian Togelius and Sebastian Risi",

note = "Funding Information: Manuscript received December 7, 2017; revised April 9, 2018, July 24, 2018, and December 7, 2018; accepted January 7, 2019. Date of publication February 13, 2019; date of current version March 17, 2020. The work of N. Justesen was supported by the Elite Research travel grant from The Danish Ministry for Higher Education and Science. (Corresponding author: Niels Justesen.) N. Justesen and S. Risi are with the IT University of Copenhagen, 2300 Copenhagen, Denmark (e-mail:, [email protected]; sebastian.risi@ gmail.com). Publisher Copyright: {\textcopyright} 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.",

year = "2020",

month = mar,

doi = "10.1109/TG.2019.2896986",

language = "English (US)",

volume = "12",

pages = "1--20",

journal = "IEEE Transactions on Games",

issn = "2475-1502",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1",

}

TY - JOUR

T1 - Deep learning for video game playing

AU - Justesen, Niels

AU - Bontrager, Philip

AU - Togelius, Julian

AU - Risi, Sebastian

N1 - Funding Information: Manuscript received December 7, 2017; revised April 9, 2018, July 24, 2018, and December 7, 2018; accepted January 7, 2019. Date of publication February 13, 2019; date of current version March 17, 2020. The work of N. Justesen was supported by the Elite Research travel grant from The Danish Ministry for Higher Education and Science. (Corresponding author: Niels Justesen.) N. Justesen and S. Risi are with the IT University of Copenhagen, 2300 Copenhagen, Denmark (e-mail:, [email protected]; sebastian.risi@ gmail.com). Publisher Copyright: © 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.

PY - 2020/3

Y1 - 2020/3

N2 - In this paper, we review recent deep learning advances in the context of how they have been applied to play different types of video games such as first-person shooters, arcade games, and real-time strategy games. We analyze the unique requirements that different game genres pose to a deep learning system and highlight important open challenges in the context of applying these machine learning methods to video games, such as general game playing, dealing with extremely large decision spaces and sparse rewards.

AB - In this paper, we review recent deep learning advances in the context of how they have been applied to play different types of video games such as first-person shooters, arcade games, and real-time strategy games. We analyze the unique requirements that different game genres pose to a deep learning system and highlight important open challenges in the context of applying these machine learning methods to video games, such as general game playing, dealing with extremely large decision spaces and sparse rewards.

KW - Algorithms

KW - Artificial intelligence

KW - Learning

KW - Machine learning algorithms

KW - Multilayer neural network

UR - http://www.scopus.com/inward/record.url?scp=85089102332&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85089102332&partnerID=8YFLogxK

U2 - 10.1109/TG.2019.2896986

DO - 10.1109/TG.2019.2896986

M3 - Article

AN - SCOPUS:85089102332

SN - 2475-1502

VL - 12

SP - 1

EP - 20

JO - IEEE Transactions on Games

JF - IEEE Transactions on Games

IS - 1

M1 - 2896986

ER -

Deep learning for video game playing

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this