AlphaZero

E40166

artificial intelligence system game‑playing program

AlphaZero is a DeepMind-developed artificial intelligence system that mastered complex games like chess, shogi, and Go through self-play reinforcement learning without human-crafted strategies.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (6)

Label	Occurrences
AlphaZero canonical	13
AlphaGo Zero	3
A general reinforcement learning algorithm that masters chess, shogi, and Go through self‑play	1
AlphaZero (shogi) in DeepMind experiments	1
DeepMind AlphaZero	1
Mastering chess and shogi by self-play with a general reinforcement learning algorithm	1

Statements (53)

Predicate	Object
instanceOf	artificial intelligence system ⓘ game‑playing program ⓘ
architectureType	deep neural network with Monte Carlo tree search ⓘ
basedOn	Monte Carlo tree search ⓘ deep learning ⓘ reinforcement learning ⓘ
contrastWith	programs relying on human expert knowledge ⓘ traditional chess engines using alpha‑beta search ⓘ
countryOfOrigin	United Kingdom ⓘ
creatorOrganizationType	AI research lab ⓘ
defeated	Elmo shogi engine ⓘ Stockfish ⓘ surface form: Stockfish 8 previous Go programs based on AlphaGo Zero ⓘ
designedFor	Go ⓘ chess ⓘ shogi ⓘ
developer	DeepMind ⓘ DeepMind ⓘ surface form: Google DeepMind
doesNotUse	endgame tablebases for search guidance ⓘ human‑crafted opening books ⓘ
evaluationFunction	learned value function ⓘ
field	artificial intelligence ⓘ computer Go ⓘ computer chess ⓘ computer shogi ⓘ machine learning ⓘ
firstPublicAnnouncementDate	2017-12-06 ⓘ
firstPublicAnnouncementYear	2017 ⓘ
gameRepresentation	board positions encoded for neural networks ⓘ
generalizationProperty	single algorithm applied to multiple games ⓘ
hardwareUsed	TPUs (via XLA integrations) ⓘ surface form: TPUs
learningObjective	maximize expected game outcome ⓘ
learningParadigm	tabula rasa learning ⓘ
notableFor	mastering Go through self‑play ⓘ mastering chess through self‑play ⓘ mastering shogi through self‑play ⓘ
outperforms	AlphaGo Zero ⓘ Elmo ⓘ Stockfish ⓘ
parentProject	AlphaGo ⓘ surface form: AlphaGo project
policyRepresentation	probability distribution over moves ⓘ
publicationTitle	AlphaZero self-linksurface differs ⓘ surface form: A general reinforcement learning algorithm that masters chess, shogi, and Go through self‑play
publishedIn	Science ⓘ
rewardSignal	game result win‑draw‑loss ⓘ
searchGuidance	policy network priors ⓘ value network evaluations ⓘ
searchTechnique	Monte Carlo tree search guided by neural networks ⓘ
trainingDataSource	self‑generated game data ⓘ
trainingMethod	self‑play ⓘ
trainingRegime	self‑play reinforcement learning without human examples ⓘ
uses	neural networks ⓘ policy network ⓘ value network ⓘ

Referenced by (20)

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

DeepMind → knownFor → AlphaZero ⓘ

DeepMind → developed → AlphaZero ⓘ

this entity surface form: AlphaGo Zero

DeepMind → developed → AlphaZero ⓘ

Demis Hassabis → knownFor → AlphaZero ⓘ

AlphaGo → successor → AlphaZero ⓘ

this entity surface form: AlphaGo Zero

AlphaGo → successor → AlphaZero ⓘ

AlphaGo → inspired → AlphaZero ⓘ

this entity surface form: AlphaGo Zero

AlphaGo → inspired → AlphaZero ⓘ

David Silver → knownFor → AlphaZero ⓘ

David Silver → notableWork → AlphaZero ⓘ

David Silver → notablePaper → AlphaZero ⓘ

this entity surface form: Mastering chess and shogi by self-play with a general reinforcement learning algorithm

AlphaStar → relatedTo → AlphaZero ⓘ

AlphaZero → publicationTitle → AlphaZero self-linksurface differs ⓘ

this entity surface form: A general reinforcement learning algorithm that masters chess, shogi, and Go through self‑play

MuZero → inspiredBy → AlphaZero ⓘ

MuZero → comparedTo → AlphaZero ⓘ

Demis → knownFor → AlphaZero ⓘ

subject surface form: Demis Hassabis

Ioannis Antonoglou → knownFor → AlphaZero ⓘ

Elmo shogi engine → usedBy → AlphaZero ⓘ

this entity surface form: DeepMind AlphaZero

Elmo shogi engine → comparedWith → AlphaZero ⓘ

this entity surface form: AlphaZero (shogi) in DeepMind experiments

AlphaGo Zero → inspiredDevelopmentOf → AlphaZero ⓘ