Deceptive games

Damien Anderson, Matthew Stephenson, Julian Togelius, Christian Salge, John Levine, Jochen Renz

Research output: Contribution to journalConference Contribution

7 Citations (Scopus)

Abstract

Deceptive games are games where the reward structure or other aspects of the game are designed to lead the agent away from a globally optimal policy. While many games are already deceptive to some extent, we designed a series of games in the Video Game Description Language (VGDL) implementing specific types of deception, classified by the cognitive biases they exploit. VGDL games can be run in the General Video Game Artificial Intelligence (GVGAI) Framework, making it possible to test a variety of existing AI agents that have been submitted to the GVGAI Competition on these deceptive games. Our results show that all tested agents are vulnerable to several kinds of deception, but that different agents have different weaknesses. This suggests that we can use deception to understand the capabilities of a game-playing algorithm, and game-playing algorithms to characterize the deception displayed by a game.
LanguageEnglish
Pages376-391
Number of pages16
JournalLecture Notes in Computer Science
Volume10784
Early online date8 Mar 2018
DOIs
Publication statusE-pub ahead of print - 8 Mar 2018

Fingerprint

Game
Deception
Video Games
Artificial intelligence
Artificial Intelligence
Optimal Policy
Reward
Series

Keywords

  • games
  • tree search
  • reinforcement learning
  • deception

Cite this

Anderson, D., Stephenson, M., Togelius, J., Salge, C., Levine, J., & Renz, J. (2018). Deceptive games. Lecture Notes in Computer Science, 10784, 376-391. https://doi.org/10.1007/978-3-319-77538-8_26
Anderson, Damien ; Stephenson, Matthew ; Togelius, Julian ; Salge, Christian ; Levine, John ; Renz, Jochen. / Deceptive games. In: Lecture Notes in Computer Science. 2018 ; Vol. 10784. pp. 376-391.
@article{efb0f4c828434f7299a0f29c137d77b1,
title = "Deceptive games",
abstract = "Deceptive games are games where the reward structure or other aspects of the game are designed to lead the agent away from a globally optimal policy. While many games are already deceptive to some extent, we designed a series of games in the Video Game Description Language (VGDL) implementing specific types of deception, classified by the cognitive biases they exploit. VGDL games can be run in the General Video Game Artificial Intelligence (GVGAI) Framework, making it possible to test a variety of existing AI agents that have been submitted to the GVGAI Competition on these deceptive games. Our results show that all tested agents are vulnerable to several kinds of deception, but that different agents have different weaknesses. This suggests that we can use deception to understand the capabilities of a game-playing algorithm, and game-playing algorithms to characterize the deception displayed by a game.",
keywords = "games, tree search, reinforcement learning, deception",
author = "Damien Anderson and Matthew Stephenson and Julian Togelius and Christian Salge and John Levine and Jochen Renz",
year = "2018",
month = "3",
day = "8",
doi = "10.1007/978-3-319-77538-8_26",
language = "English",
volume = "10784",
pages = "376--391",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",

}

Anderson, D, Stephenson, M, Togelius, J, Salge, C, Levine, J & Renz, J 2018, 'Deceptive games' Lecture Notes in Computer Science, vol. 10784, pp. 376-391. https://doi.org/10.1007/978-3-319-77538-8_26

Deceptive games. / Anderson, Damien; Stephenson, Matthew; Togelius, Julian; Salge, Christian; Levine, John; Renz, Jochen.

In: Lecture Notes in Computer Science, Vol. 10784, 08.03.2018, p. 376-391.

Research output: Contribution to journalConference Contribution

TY - JOUR

T1 - Deceptive games

AU - Anderson, Damien

AU - Stephenson, Matthew

AU - Togelius, Julian

AU - Salge, Christian

AU - Levine, John

AU - Renz, Jochen

PY - 2018/3/8

Y1 - 2018/3/8

N2 - Deceptive games are games where the reward structure or other aspects of the game are designed to lead the agent away from a globally optimal policy. While many games are already deceptive to some extent, we designed a series of games in the Video Game Description Language (VGDL) implementing specific types of deception, classified by the cognitive biases they exploit. VGDL games can be run in the General Video Game Artificial Intelligence (GVGAI) Framework, making it possible to test a variety of existing AI agents that have been submitted to the GVGAI Competition on these deceptive games. Our results show that all tested agents are vulnerable to several kinds of deception, but that different agents have different weaknesses. This suggests that we can use deception to understand the capabilities of a game-playing algorithm, and game-playing algorithms to characterize the deception displayed by a game.

AB - Deceptive games are games where the reward structure or other aspects of the game are designed to lead the agent away from a globally optimal policy. While many games are already deceptive to some extent, we designed a series of games in the Video Game Description Language (VGDL) implementing specific types of deception, classified by the cognitive biases they exploit. VGDL games can be run in the General Video Game Artificial Intelligence (GVGAI) Framework, making it possible to test a variety of existing AI agents that have been submitted to the GVGAI Competition on these deceptive games. Our results show that all tested agents are vulnerable to several kinds of deception, but that different agents have different weaknesses. This suggests that we can use deception to understand the capabilities of a game-playing algorithm, and game-playing algorithms to characterize the deception displayed by a game.

KW - games

KW - tree search

KW - reinforcement learning

KW - deception

U2 - 10.1007/978-3-319-77538-8_26

DO - 10.1007/978-3-319-77538-8_26

M3 - Conference Contribution

VL - 10784

SP - 376

EP - 391

JO - Lecture Notes in Computer Science

T2 - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -

Anderson D, Stephenson M, Togelius J, Salge C, Levine J, Renz J. Deceptive games. Lecture Notes in Computer Science. 2018 Mar 8;10784:376-391. https://doi.org/10.1007/978-3-319-77538-8_26