List your company

Game Playing on Yahtzee Full-game 100,000 evaluation games

254.59Mean ScoreDP Optimal
187.25204.7325222.215239.6975Dec 18, 2025

Evaluation results

MethodLinks
DP Optimal2025.12254.59—68.1233.7448.37
A2CTraining Budget=1M games2025.12241.783,230.8624.9334.0536.87
A2CTraining Budget=250K games2025.12230.382,503.8311.3731.0827.82
PPOlambda=0.3, k=4, Training Budget=250K games2025.12230.22,345.519.7124.8728.38
PPOlambda=0.3, k=5, Training Budget=50k games2025.12204.54860.022.496.546.08
REINFORCEmode=full-game, Training Budget=1M games2025.12203.731,265.836.4410.619.22
REINFORCEmode=single-turn, Training Budget=500K games2025.12201.481,3660.8919.639.34
REINFORCEmode=full-game, Training Budget=250K games2025.12189.84812.242.831.72.06