David Kruger 854772a63d Improve UCB1-Tuned performance
We calculate the sample variance of the rewards online storing the value
in the node. This greatly reduces the amount of summations that need to
be done to calculate the variance during the selection phase.

While this burdens other selection algorithms, the cost is not
substantial.
2025-06-29 18:43:52 -07:00
2025-06-27 22:25:56 -07:00
2025-06-29 18:43:52 -07:00
2025-06-12 11:13:49 -07:00
2025-06-27 22:25:56 -07:00
2025-06-27 22:25:56 -07:00
2025-06-12 11:11:51 -07:00
2025-06-12 11:11:51 -07:00

rustic_mcts

An extensible implementation of Monte Carlo Tree Search (MCTS) using an arena allocator.

Description
An extensible implementation of Monte Carlo Tree Search (MCTS) using an arena allocator.
Readme MIT 106 KiB
Languages
Rust 100%