Monte Carlo tree search

The technique

Instead of examining every branch to a fixed depth, spend each unit of thinking where past playouts say it is most worth spending, and grow the tree lopsided.

Read the primer

Approaches using this technique

3 approaches, featured pages first. Each card is one theory of how to choose a column and what happened when it was tried.

Monte Carlo tree searchfeatured

Observable-state MCTS

Grow the look-ahead by simulation, but only ever let the search know what a player would actually be able to see.

rejectedrecorded

Monte Carlo tree search

PUCT

A complete but unrun tree search that starts from a prior about which columns are worth trying.

Monte Carlo tree search

TypeScript MCTS lab

A small Monte Carlo tree search you can run in a terminal and watch one decision at a time.

rejected

Agent contextApproach directories in this group, by family

tree-search

approaches/tree-search/mcts
approaches/tree-search/observable-mcts
approaches/tree-search/puct