causal
MCTS gives a strictly better label per move
MCTS acts as a policy-improvement operator: for every action taken in a game (whether the game was won or lost), the search provides a strictly better action/distribution to imitate, giving one low-variance supervision target per move rather than a single noisy win/loss signal for the whole trajectory.
causalpending
Speaker
Eric JangEvidence Quote
“you have one supervision target for every single action. The variance of your learning signal is very low compared to the alternative naive RL thing.”
Created: 6/13/2026, 3:37:56 AM
Your notes
Loading notes...