Monte Carlo tree search

E205830

Monte Carlo method decision-making technique game tree search method heuristic algorithm search algorithm

Monte Carlo tree search is a heuristic search algorithm that uses random sampling of game states to build and explore a search tree, enabling strong decision-making in complex domains like Go and other board games.

Try in SPARQL Jump to: Surface forms Disambiguation Statements Elicitation Referenced by

All labels observed (7)

Label	Occurrences
Monte Carlo tree search canonical	9
Monte Carlo Tree Search	3
Nested Monte Carlo Search	1
Parallel MCTS	1
Rapid Action Value Estimation	1
Sample-based Search for Optimal Planning in Markov Decision Processes	1
Upper Confidence bounds applied to Trees	1

How this entity was disambiguated

This entity first appeared as the object of triple T1844415 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.

NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07

Target entity: Monte Carlo tree search
Context triple: [David Silver, researchInterest, Monte Carlo tree search]

A. MuZero
MuZero is a DeepMind reinforcement learning algorithm that learns to plan and master complex games like Go, chess, and Atari without being given the rules in advance.
B. AlphaZero
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.
C. AlphaGo
AlphaGo is an artificial intelligence program developed by DeepMind that became famous for defeating world champion Go players using deep neural networks and reinforcement learning.
D. Atari deep Q-network
The Atari deep Q-network is a pioneering deep reinforcement learning system that learned to play a wide range of Atari 2600 video games directly from raw pixels at human-level or better performance.
E. AlphaStar
AlphaStar is a DeepMind-created artificial intelligence system that achieved grandmaster-level performance in the real-time strategy game StarCraft II.
F. None of above. chosen
G. Unsure - the case is ambiguous/there is not enough information to decide.

NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07

Target entity: Monte Carlo tree search
Target entity description: Monte Carlo tree search is a heuristic search algorithm that uses random sampling of game states to build and explore a search tree, enabling strong decision-making in complex domains like Go and other board games.

A. MuZero
MuZero is a DeepMind reinforcement learning algorithm that learns to plan and master complex games like Go, chess, and Atari without being given the rules in advance.
B. AlphaZero
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.
C. AlphaGo
AlphaGo is an artificial intelligence program developed by DeepMind that became famous for defeating world champion Go players using deep neural networks and reinforcement learning.
D. Atari deep Q-network
The Atari deep Q-network is a pioneering deep reinforcement learning system that learned to play a wide range of Atari 2600 video games directly from raw pixels at human-level or better performance.
E. AlphaStar
AlphaStar is a DeepMind-created artificial intelligence system that achieved grandmaster-level performance in the real-time strategy game StarCraft II.
F. None of above. chosen

Statements (49)

Predicate	Object
instanceOf	Monte Carlo method ⓘ decision-making technique ⓘ game tree search method ⓘ heuristic algorithm ⓘ search algorithm ⓘ
abbreviation	MCTS ⓘ
advantage	requires minimal domain knowledge ⓘ scales well with additional computation time ⓘ
appliedIn	artificial intelligence research ⓘ board games ⓘ combinatorial optimization ⓘ computer Go ⓘ general game playing ⓘ planning problems ⓘ puzzles ⓘ real-time video games ⓘ robotics ⓘ
canUse	domain-specific heuristics ⓘ neural network policy priors ⓘ neural network value functions ⓘ
developedInContextOf	computer Go research ⓘ
goal	select strong actions in complex decision spaces ⓘ
hasComponent	backpropagation phase ⓘ expansion phase ⓘ selection phase ⓘ simulation phase ⓘ
input	current game state ⓘ
notableUse	strong Go programs before deep learning era ⓘ
oftenUses	upper confidence bounds applied to trees ⓘ
optimizes	exploration–exploitation trade-off ⓘ
output	estimated value of positions ⓘ recommended move ⓘ
property	anytime algorithm ⓘ asymmetric tree growth ⓘ can be interrupted at any time with best-so-far move ⓘ can handle large branching factors ⓘ does not require full game tree enumeration ⓘ
relatedTo	alpha–beta pruning ⓘ minimax search ⓘ multi-armed bandit problem ⓘ upper confidence bound algorithm ⓘ
requires	rollout strategy ⓘ simulation policy ⓘ
usedIn	AlphaGo search procedure ⓘ AlphaZero search procedure ⓘ MuZero search procedure ⓘ
uses	Monte Carlo simulations ⓘ playout simulations ⓘ random sampling of game states ⓘ

How these facts were elicited

Referenced by (17)

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

David Silver → researchInterest → Monte Carlo tree search ⓘ

AlphaZero → basedOn → Monte Carlo tree search ⓘ

MuZero → basedOn → Monte Carlo Tree Search ⓘ