Offline Reinforcement Learning on hopper medium-replay benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on hopper medium-replay

113Normalized ScoreProposed

Evaluation Results

Method	Links
Proposed2023.10		113
Anti-exploration Method2026.02		103.1
ATAC2023.10		102.5
Q-ALIGN DT2026.05		102.2
QT2026.05		102
DMG2026.02		101.9
SAC-RND2026.02		100.5
QCS2026.05		100.4
O-DICE2026.02		99.9
GPC-SAC2026.02		97.5
RADT2026.05		95.7
DM2026.05		95.4
IQL2023.10		94.7
IQL2026.02		94.7
DC2026.05		94.2
Exp. Weight2021.06		94.1
LSDT2026.05		93.9
CGDT2026.05		93.4
IQL2026.05		92.1
COMBO2023.10		89.5
I-TAP2026.02		84.43
L-MAP2026.02		83.99
DT2026.05		82.7
Easy BCQ2021.06		77.3
I-TAP2026.02		75.22
Rev. KL Reg.2021.06		71
L-MAP2026.02		70.06
DT2024.02		63.7
TD3+BC2026.05		60.9
I-TAP2026.02		56.96
QDT2024.02		52.1
DT2024.02		50.8
L-MAP2026.02		50.79
CQL2024.02		48.6
Fu et al.2021.06		48.6
TD3-CVAE2026.02		46.7
TD3+BC2023.10		44.4
DT2026.02		39.22
QDT2024.02		38.7
OptiDICE2023.10		36.4
CQL2026.02		35.9
BEAR2023.10		33.7
BCQ2023.10		33.1
DT2026.02		33.03
CQL2023.10		32.6
DT2026.02		30.17
BC2021.06		21.2
CQL2024.02		7.7
DVDF-IGDF2025.12		7.4
DVDF-OTDF2025.12		5.6
OTDF2025.12		4.6
IGDF2025.12		4.1
BOSA2025.12		3.7
DARA2025.12		3.5
IQL2025.12		2.3