Offline Reinforcement Learning on HalfCheetah random benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on HalfCheetah random

33.8Normalized ScoreAnti-exploration Method

Evaluation Results

Method	Links
Anti-exploration Method2026.02		33.8
GPC-SAC2026.02		31
SAC-RND2026.02		29
DMG2026.02		28.8
TD3-CVAE2026.02		28.6
CQL2026.02		28.3
IQL2026.02		11
BOSA2025.12		2.2
DVDF-OTDF2025.12		2.2
DARA2025.12		2
OTDF2025.12		2
IQL2025.12		0
IGDF2025.12		0
DVDF-IGDF2025.12		0