Share your thoughts, 1 month free Claude Pro on usSee more

Reinforcement Learning on MiniHack Corridor-5

1Mean Return

PPO-RNN

Updated 1mo ago

Evaluation Results

Method	Links
PPO-RNN 2026.03		1	0.29	52
PPO-RNN 2026.03		1	0.29	52
PPO-RNN 2026.03		1	0.16	52.1
PPO-RNN 2026.03		1	0.21	52
PPO-RNN 2026.03		1	0.21	52
PPO-RNN 2026.03		1	0.21	53.8
PPO-RNN 2026.03		1	0.21	53.8
PPO-Hybrid 2026.03		1	0.84	94.6
PPO-Hybrid 2026.03		1	0.84	94.6
PPO-Hybrid 2026.03		1	0.63	80.7
PPO-Hybrid 2026.03		1	0.76	92
PPO-Hybrid 2026.03		1	0.76	92
PPO-Hybrid 2026.03		1	0.65	88
PPO-Hybrid 2026.03		1	0.65	88
PPO-MLP 2026.03		0.9	-	161.3
PPO-MLP 2026.03		0.9	-	161.3
PPO-MLP 2026.03		0.9	-	118.3
PPO-MLP 2026.03		0.9	-	113.6
PPO-MLP 2026.03		0.9	-	113.6
PPO-MLP 2026.03		0.9	-	123.6
PPO-MLP 2026.03		0.9	-	123.6
PPO-Hybrid 2026.03		0.3	0.63	80.7
PPO-RNN 2026.03		0.2	0.16	52.1
PPO-MLP 2026.03		0.1	-	118.3