List your company

Offline Reinforcement Learning on D4RL Adroit (expert, human)

46.1Adroit Door Return (Human)KFC++
-2.15610.37222.935.428Jun 7, 2021Dec 4, 2021Jun 2, 2022Nov 29, 2022May 28, 2023Nov 24, 2023May 23, 2024

Evaluation results

MethodLinks
KFC++2021.1146.1————0.2609.4—68.44.25.6-0.1—
KFC2021.1144.1————0.261.37—71.333.6-0.1—
S4RL-(Adv)Augmentation=Adversarial2021.1135.3————0.251.26.3—58.22.90.8-0.1—
S4RL-(N)Augmentation=Normal2021.1127————0.244.45.9—57.12.72.1-0.1—
EPQ2024.0513.2————0.383.93.9—91.822.85.80.1221.8
CABI+IQL2022.0611.5105.7107.4129.1127.90.4724.8604.142.22.40.8-0.1—
CQL2021.119.9————0.237.54.4—39.22.10.4-0.1—
CQL2022.069.9101.59510786.70.237.54.4483.839.22.10.4-0.1—
CQLsource=reproduced2024.059.9————0.237.54.4—39.22.10.4-0.193.6
CQLrandom_seeds=32021.069.1————0.3555.82.1——————
CQLsource=original paper2024.059.1————0.455.82.1—40.35.73.5-0.1116.8
ICQrandom_seeds=32021.066.4103.9109.5123.8128.31.591.32——————
Onestep2024.055.4————1.971.81.2—602.10.4-0.1142.7
MISA2024.055.2————0.188.18.1—58.62.20.5-0.1162.7
IQL2024.054.3————0.171.51.4—37.32.11.6-0.2118.1
IQL2022.063.3105.2105.6118.4121.7068.72.3562.535.21.60.7-0.2—
MCQ2024.052.3————0.168.50.3—49.41.41.30123.3
UWAC2022.062.1104.1105.6111.9110.60.521.71.1490.833.10.40-0.3—
BC2024.052————0.163.91.2—370.60-0.3104.5
BCrandom_seeds=32021.060.5101.2101.385.1125.6034.41.5——————
BC2022.060.534.9101.385.1125.6034.41.5440.856.90.8-0.1-0.1—
AWRrandom_seeds=32021.060.4102.991.511139012.31.2——————
AWR2022.060.4102.991.511139012.31.2386.5280.40-0.2—
BCQrandom_seeds=32021.0609941.6114.9107.2-0.168.90.5——————
TD3+BC2024.050————0.164.81.8—490.20-0.2115.7
MOPO2022.06-0.2-0.2-0.2-0.60.3-0.39.70.26.4-2.10.2-0.1-0.3—
BRAC-prandom_seeds=32021.06-0.3-0.3-0.3-3.50.3-0.38.10.3——————
BEAR2022.06-0.3103.498.6105.9127.3-0.3-10.3460.326.50.3-0.1-0.3—
COMBO2022.06-0.34.917.211.50.3-0.327.70.259.1-2.40.40-0.1—