Reinforcement Learning on Kangaroo Atari (classical) benchmark leaderboard · SOTA2 Research

Reinforcement Learning on Kangaroo Atari (classical)

131,842RewardH2RL++

Evaluation Results

Method	Links
H2RL++2026.03		131,842
PPO2026.03		14,592
hPPO2026.03		10,601
H2RL2026.03		5,148
H2RL+2026.03		2,754
hReason2026.03		1,701
BlendRL2026.03		1,482
NUDGE2026.03		404
hDQN2026.03		322
Option-critic2026.03		61