x̄ - > Infinite Tic-Tac-Toe — RL Policy Animation

Infinite Tic-Tac-Toe — RL Policy Animation

A lightweight demonstration of a heuristic RL-inspired policy. X tends to build threats; O spreads and blocks.

Speed:

Grid: 21×21 31×31 41×41

Policy Notes

X threat-seeker (open-three bias)
O reactive spreader / blocker

This uses heuristic scoring to imitate RL behavior (distance features, threat windows, local density, softmax selection).

Tip: click Step to see feature calculations in the browser console.