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

Offline Reinforcement Learning on D4RL AntMaze

99.8AntMaze Umaze Return

KFC++

Updated 24d ago

Evaluation Results

Method
KFC++ 2021.11	99.8	-	91.1	63.1	90.5	25.6	34	-	-	-
GSFlow (Ours) 2026.03	99.6	-	93.2	77.2	75.5	86.5	84.8	-	86.1	-
EPQ 2024.05	99.4	-	78.3	85	86.7	40	36.7	426.1	-	-
ReBRAC 2026.03	99	-	89	86.7	83	59	51	-	77.62	-
MPCwDWM 2026.03	99	-	90	94.4	94	67	66	-	85.07	-
MCQ 2024.05	98.3	-	80	52.5	37.5	2.5	7.5	278.3	-	-
ReBRAC 2026.03	98	-	84	90	84	52	64	-	78.7	-
ReBRAC 2026.03	97.75	-	83.5	89.5	83.5	52.25	64	-	78.42	470.5
SRPO 2026.03	97	-	82	81	75	54	54	-	73.8	-
KFC 2021.11	96.9	-	91.2	60	87.1	24.8	33.1	-	-	-
FQL 2026.03	96	-	89	78	71	84	83	-	83.5	-
FQL 2026.03	96	-	89	78	71	84	83	-	83.5	501
SSAR 2026.03	94.7	-	65.1	59.8	59.6	35.5	26.7	-	56.9	341.4
S4RL-(Adv) 2021.11	94.1	-	88	61.6	82.3	25.1	26.2	-	-	-
IDQL 2026.03	94	-	80	84	85	64	68	-	79.2	-
FBRAC 2026.03	94	-	82	77	77	32	20	-	63.7	-
IDQL 2026.03	94	-	80.2	84.5	84.8	63.5	67.9	-	79.15	474.9
SPOT 2022.02	93.5	-	65.6	74.7	79.1	35.3	36.3	359.6	-	-
Diffusion-QL 2026.03	93.4	-	66.2	76.6	78.6	46.4	56.6	-	69.63	417.8
SPQR-CQL-Min 2024.01	93.3	-	80	80	-	-	-	-	84.4	-
CQL 2026.03	92.75	-	37.25	65.75	67.25	20.75	20.5	-	50.71	304.25
DMG 2024.11	92.4	-	75.4	80.2	77.2	55.4	58.8	439.4	-	-
MISA 2024.05	92.3	-	89.1	63	62.8	17.5	23.4	348.1	-	-
IFQL 2026.03	92	-	62	56	60	55	64	-	64.8	-
S4RL-(N) 2021.11	91.3	-	87.8	61.9	78.1	24.4	27	-	-	-
FAWAC 2026.03	90	-	55	52	44	10	16	-	44.5	-
IQL 2022.02	89.6	-	65.6	76.4	72.8	42	46	392.4	-	-
IQL 2024.11	89.6	-	65.6	76.4	72.8	42	46	392.4	-	-
IQL 2024.05	87.5	-	62.2	71.2	70	39.6	47.5	378	-	-
GEM 2026.03	83	-	79.33	78.33	82.67	37.33	37	-	66.28	397.66
CQL 2022.02	82.6	-	10.2	59	46.6	16.4	3.2	218	-	-
CQL 2024.11	82.6	-	10.2	59	46.6	16.4	3.2	218	-	-
CQL 2024.05	80.4	-	56.3	67.5	62.5	35	13.3	315	-	-
AWAC 2024.11	80	-	52	0	0.2	0	0	132.2	-	-
BCQ 2022.02	78.9	-	55	0	0	6.7	2.2	142.8	-	-
BCQ 2024.11	78.9	-	55	0	0	6.7	2.2	142.8	-	-
TD3+BC 2024.05	78.6	-	71.4	10.6	3	0.2	0	163.8	-	-
IQL 2026.03	77	-	54	66	74	42	30	-	57.2	-
IQL 2026.03	77	-	54.25	65.75	73.75	42	30.25	-	57.17	343
CQL 2021.11	74	-	84	61.2	53.7	15.8	14.9	-	-	-
CQL-Min 2024.01	74	-	84	61.2	-	-	-	-	73.1	-
CQL 2024.05	74	-	84	61.2	53.7	15.8	14.9	303.6	-	-
BEAR 2022.02	73	-	61	0	8	0	0	142	-	-
TD3+BC 2022.02	73	-	47	0	0.2	0	0	120.2	-	-
BEAR 2024.01	73	-	61	0	-	-	-	-	44.7	-
BEAR 2024.11	73	-	61	0	8	0	0	142	-	-
TD3BC 2024.11	73	-	47	0	0.2	0	0	120.2	-	-
Onestep 2024.05	72.5	-	75	5	5	2.5	2.5	162.5	-	-
TD3+BC 2026.03	70.75	-	44.75	0.25	0.25	0	0	-	19.33	116
BC 2024.11	66.8	-	56.8	0	0	0	0	123.6	-	-
CAC 2026.03	66	-	66	49	0	0	0	-	30.2	-
BC 2024.01	65	-	55	0	-	-	-	-	40	-
10% BC 2024.05	62.8	-	50.2	5.4	9.8	0	6	134.2	-	-
PLAS 2022.02	62	-	45.4	31.4	20.6	2.2	3	164.6	-	-
DT 2026.03	57	-	51.75	0	0	0	0	-	18.12	108.75
AWAC 2026.03	56.75	-	54.75	0	0	0	0	-	18.58	111.5
BC 2026.03	55	-	47	0	1	0	0	-	17.2	-
BC 2024.05	54.6	-	45.6	0	0	0	0	100.2	-	-
Decision Transformer 2022.02	54.2	-	41.2	0	0	0	0	95.4	-	-
DT 2024.11	54.2	-	41.2	0	0	0	0	95.4	-	-
OneStep 2024.11	54	-	57.8	0	0.6	0	0.2	112.6	-	-
BC 2022.02	49.2	-	41.8	0.4	0.2	0	0	91.6	-	-
SAC-Min 2024.01	0	-	0	0	-	-	-	-	0	-
SAC-N 2026.03	0	-	0	0	0	0	0	-	0	0
EDAC 2026.03	0	-	0	0	0	0	0	-	0	0
Original 2025.11	-	57.3	-	-	-	-	-	-	-	-
ASTRO 2025.11	-	70.4	-	-	-	-	-	-	-	-
DiffStitch 2025.11	-	65.3	-	-	-	-	-	-	-	-
SynthER 2025.11	-	63.5	-	-	-	-	-	-	-	-
Original 2025.11	-	78.2	-	-	-	-	-	-	-	-
ASTRO 2025.11	-	88.6	-	-	-	-	-	-	-	-
DiffStitch 2025.11	-	85	-	-	-	-	-	-	-	-
SynthER 2025.11	-	79	-	-	-	-	-	-	-	-