The Machine Reconnaissance Blind Chess Tournament of NeurIPS 2022

Ryan W. Gardner; Gino Perrotta; Anvay Shah; Shivaram Kalyanakrishnan; Kevin A. Wang; Gregory Clark; Timo Bertram; Johannes Fürnkranz; Martin Müller; Brady P. Garrison; Prithviraj Dasgupta; Saeid Rezaei

The Machine Reconnaissance Blind Chess Tournament of NeurIPS 2022

Ryan W. Gardner
, Gino Perrotta
, Anvay Shah
, Shivaram Kalyanakrishnan
, Kevin A. Wang
, Gregory Clark
, Timo Bertram
, Johannes Fürnkranz
, Martin Müller
, Brady P. Garrison
, Prithviraj Dasgupta
, Saeid Rezaei

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

Abstract

Reconnaissance Blind Chess is a game that plays like regular chess but rather than continuously observing the entire board, each player can only momentarily and privately observe selected board regions. It has imperfect information and little common knowledge. The Johns Hopkins University Applied Physics Laboratory (the game’s creator) and several partners organized the third NeurIPS machine Reconnaissance Blind Chess competition in 2022 to bring people together to attempt to tackle research challenges presented by the game. 18 bots played each other in 9,180 games (60 matches per bot pair) over 4 days. The top bot exceeded the performance of all of last year’s bots yet a practical, sound (unexploitable) algorithm remains unknown.

Original language	English
Pages (from-to)	119-132
Number of pages	14
Journal	Proceedings of Machine Learning Research
Volume	220
Publication status	Published - 2023
Event	36th Annual Conference on Neural Information Processing Systems - Competition Track, NeurIPS 2022 - Virtual, Online, United States Duration: 28 Nov 2022 → 9 Dec 2022

Keywords

common knowledge
imperfect information
reconnaissance blind chess
reinforcement learning

Cite this

@article{7376821d21ef42a7adfa0a9d4c16a702,

title = "The Machine Reconnaissance Blind Chess Tournament of NeurIPS 2022",

abstract = "Reconnaissance Blind Chess is a game that plays like regular chess but rather than continuously observing the entire board, each player can only momentarily and privately observe selected board regions. It has imperfect information and little common knowledge. The Johns Hopkins University Applied Physics Laboratory (the game{\textquoteright}s creator) and several partners organized the third NeurIPS machine Reconnaissance Blind Chess competition in 2022 to bring people together to attempt to tackle research challenges presented by the game. 18 bots played each other in 9,180 games (60 matches per bot pair) over 4 days. The top bot exceeded the performance of all of last year{\textquoteright}s bots yet a practical, sound (unexploitable) algorithm remains unknown.",

keywords = "common knowledge, imperfect information, reconnaissance blind chess, reinforcement learning",

author = "Gardner, \{Ryan W.\} and Gino Perrotta and Anvay Shah and Shivaram Kalyanakrishnan and Wang, \{Kevin A.\} and Gregory Clark and Timo Bertram and Johannes F{\"u}rnkranz and Martin M{\"u}ller and Garrison, \{Brady P.\} and Prithviraj Dasgupta and Saeid Rezaei",

note = "Publisher Copyright: {\textcopyright} 2023 R.W. Gardner et al.; 36th Annual Conference on Neural Information Processing Systems - Competition Track, NeurIPS 2022 ; Conference date: 28-11-2022 Through 09-12-2022",

year = "2023",

language = "English",

volume = "220",

pages = "119--132",

journal = "Proceedings of Machine Learning Research",

issn = "2640-3498",

publisher = "ML Research Press",

}

TY - JOUR

T1 - The Machine Reconnaissance Blind Chess Tournament of NeurIPS 2022

AU - Gardner, Ryan W.

AU - Perrotta, Gino

AU - Shah, Anvay

AU - Kalyanakrishnan, Shivaram

AU - Wang, Kevin A.

AU - Clark, Gregory

AU - Bertram, Timo

AU - Fürnkranz, Johannes

AU - Müller, Martin

AU - Garrison, Brady P.

AU - Dasgupta, Prithviraj

AU - Rezaei, Saeid

PY - 2023

Y1 - 2023

N2 - Reconnaissance Blind Chess is a game that plays like regular chess but rather than continuously observing the entire board, each player can only momentarily and privately observe selected board regions. It has imperfect information and little common knowledge. The Johns Hopkins University Applied Physics Laboratory (the game’s creator) and several partners organized the third NeurIPS machine Reconnaissance Blind Chess competition in 2022 to bring people together to attempt to tackle research challenges presented by the game. 18 bots played each other in 9,180 games (60 matches per bot pair) over 4 days. The top bot exceeded the performance of all of last year’s bots yet a practical, sound (unexploitable) algorithm remains unknown.

AB - Reconnaissance Blind Chess is a game that plays like regular chess but rather than continuously observing the entire board, each player can only momentarily and privately observe selected board regions. It has imperfect information and little common knowledge. The Johns Hopkins University Applied Physics Laboratory (the game’s creator) and several partners organized the third NeurIPS machine Reconnaissance Blind Chess competition in 2022 to bring people together to attempt to tackle research challenges presented by the game. 18 bots played each other in 9,180 games (60 matches per bot pair) over 4 days. The top bot exceeded the performance of all of last year’s bots yet a practical, sound (unexploitable) algorithm remains unknown.

KW - common knowledge

KW - imperfect information

KW - reconnaissance blind chess

KW - reinforcement learning

UR - https://www.scopus.com/pages/publications/85179123229

M3 - Article

AN - SCOPUS:85179123229

SN - 2640-3498

VL - 220

SP - 119

EP - 132

JO - Proceedings of Machine Learning Research

JF - Proceedings of Machine Learning Research

T2 - 36th Annual Conference on Neural Information Processing Systems - Competition Track, NeurIPS 2022

Y2 - 28 November 2022 through 9 December 2022

ER -

The Machine Reconnaissance Blind Chess Tournament of NeurIPS 2022

Abstract

Keywords

Other files and links

Fingerprint

Cite this