AlphaGo Zero

E229130

Go-playing computer program artificial intelligence system computer program

AlphaGo Zero is DeepMind's advanced artificial intelligence program that learned to play the board game Go at superhuman level entirely through self-play without human data.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (3)

Label	Occurrences
AlphaGo Zero canonical	3
Mastering the game of Go without human knowledge	2
Mastering the game of Go with deep neural networks and tree search	1

Statements (52)

Predicate	Object
instanceOf	Go-playing computer program ⓘ artificial intelligence system ⓘ computer program ⓘ
achievedLevel	superhuman performance in Go ⓘ
architecture	single neural network for policy and value ⓘ
basedOn	Monte Carlo tree search ⓘ convolutional neural network ⓘ deep learning ⓘ reinforcement learning ⓘ
countryOfOrigin	United Kingdom ⓘ
describedInPublication	AlphaGo Zero self-linksurface differs ⓘ surface form: Mastering the game of Go without human knowledge
developer	DeepMind ⓘ DeepMind ⓘ surface form: DeepMind Technologies
differsFrom	AlphaGo ⓘ AlphaGo ⓘ surface form: AlphaGo Lee AlphaGo ⓘ surface form: AlphaGo Master
doesNotUse	human expert games ⓘ separate policy network ⓘ separate value network ⓘ
fieldOfApplication	computer Go ⓘ
fieldOfStudy	artificial intelligence ⓘ machine learning ⓘ reinforcement learning ⓘ
hasAuthor	Adrià Puigdomènech Badia ⓘ surface form: Adrià Puigdomènech Aja Huang ⓘ Arthur Guez ⓘ David Silver ⓘ Demis Hassabis ⓘ Dominik Grewe ⓘ Ioannis Antonoglou ⓘ John Nham ⓘ Julian Schrittwieser ⓘ Karen Simonyan ⓘ Koray Kavukcuoglu ⓘ Lucas Baker ⓘ Marc Lanctot ⓘ Matthew Lai ⓘ Pushmeet Kohli ⓘ Sander Dieleman ⓘ Thomas Hubert ⓘ Thore Graepel ⓘ Timothy Lillicrap ⓘ
inspiredDevelopmentOf	AlphaZero ⓘ
notableAchievement	surpassed performance of previous AlphaGo versions ⓘ
playsGame	Go ⓘ
publicationDate	2017-10-18 ⓘ
publicationVenue	Nature ⓘ
trainingObjective	maximize probability of winning Go games ⓘ
trainingStartState	random play ⓘ
usesTrainingData	no human game data ⓘ
usesTrainingMethod	self-play ⓘ tabula rasa learning ⓘ

Referenced by (6)

Full triples — surface form annotated when it differs from this entity's canonical label.

David Silver → notableWork → AlphaGo Zero ⓘ

AlphaZero → outperforms → AlphaGo Zero ⓘ

MuZero → inspiredBy → AlphaGo Zero ⓘ

Ioannis Antonoglou → coAuthorOf → AlphaGo Zero ⓘ

this entity surface form: Mastering the game of Go with deep neural networks and tree search

Ioannis Antonoglou → coAuthorOf → AlphaGo Zero ⓘ

this entity surface form: Mastering the game of Go without human knowledge

AlphaGo Zero → describedInPublication → AlphaGo Zero self-linksurface differs ⓘ

this entity surface form: Mastering the game of Go without human knowledge