List your company
ResearchFormerly WizWand

Reinforcement Learning on Kangaroo Atari (classical)

131,842RewardH2RL++
-5,210.2430,370.6365,951.5101,532.37Mar 6, 2026
Updated 7mo ago

Evaluation results

MethodLinks
H2RL++2026.03131,842
PPO2026.0314,592
hPPO2026.0310,601
H2RL2026.035,148
H2RL+2026.032,754
hReason2026.031,701
BlendRL2026.031,482
NUDGE2026.03404
hDQN2026.03322
Option-critic2026.0361