Multi-Objective Reinforcement Learning on Maze benchmark leaderboard · SOTA2 Research

Multi-Objective Reinforcement Learning on Maze

223.55Mean Episode Reward (MER)RANDOM

Evaluation Results

Method	Links
RANDOM2026.03		223.55	0
MER-PPO2026.03		85.55	0.05
Dense Oracle2026.03		40.33	62.92
DPI2026.03		30.16	59.04
DPI-PPO2026.03		30.16	59.04
DPI2026.03		27.35	42.94
RS2026.03		23.66	0.01
FIXED2026.03		16.15	1.12
SR-PPO2026.03		15.28	61.13
ENVELOPE2026.03		10.36	0.01
HEURISTIC2026.03		3.65	0