Offline Reinforcement Learning on D4RL Locomotion (Medium, Medium-Replay, Medium-Expert v2)

66.8Score (HalfCheetah, Medium)RORL

Evaluation results

Method
RORL2024.10	66.8	104.8	102.4	61.9	102.8	90.4	107.8	112.7	121.2	870.8
EDAC2024.10	65.9	101.6	92.5	61.3	101	87.1	106.3	110.7	114.7	841.1
GTA (with MCQ)2024.05	63.76	66.75	91.22	54.17	102.19	104.24	80	99.27	111.63	85.91
MCQ with SynthER2024.05	63.57	65.61	92.31	58.34	103.54	98.02	45.64	106.02	112.22	82.81
MCQ2024.05	60.98	73.97	89.7	54.09	101.86	92.19	80.51	92.85	114.55	84.52
MCQ with S4RL2024.05	60.93	73.95	90.91	53.32	100.62	89.9	79.02	83.91	102.62	81.69
SCAS-ht2024.10	58.5	102.5	90.8	52.9	101.6	88.1	91.7	109.7	110.8	806.6
GTA (with TD3BC)2024.05	57.84	69.57	86.69	50.04	89.31	93.82	93.13	110.4	110.86	84.63
GTA (with IQL)2024.05	54.82	77.46	84.4	46.89	102.11	93.37	95.3	107.78	112.87	86.11
GTA (with CQL)2024.05	54.14	74.8	80.4	51.36	98.88	91.57	94.93	110.9	110.44	85.27
DQL2024.10	51.1	90.5	87	47.8	101.3	95.5	96.8	111.1	110.1	791.2
IQL with SynthER2024.05	49.76	69.21	80.15	46.91	102.97	90.63	91.9	94.08	112.12	81.97
TD3BC with SynthER2024.05	49.16	63.7	85.43	45.57	78.81	90.67	85.47	98.99	109.95	78.64
TD3BC with S4RL2024.05	48.74	59.34	84.63	44.53	67.39	83.42	90.78	106.1	110.21	77.24
IQL2024.05	48.65	66.35	84.34	43.35	95.76	69.6	94.57	91.69	112.37	78.52
IQL with S4RL2024.05	48.58	65.06	84.58	43.57	86.72	70.33	94.22	99.82	112.29	78.35
TD3BC2024.05	48.42	61.04	84.58	44.64	65.69	84.11	89.48	104.08	110.23	76.92
SQL2024.10	48.3	75.5	84.2	44.8	99.7	81.2	94	111.8	110	749.5
DW+IQL2024.10	47.7	62.5	80.8	44.6	79.7	65.1	93.7	81	109.7	664.8
CQL with SynthER2024.05	47.21	64.65	81.91	46.03	92.06	86.62	95.29	107.66	109.36	81.2
EQL2024.10	47.2	74.6	83.2	44.5	98.1	76.6	90.6	105.5	110.2	730.5
CQL with S4RL2024.05	47	62.72	81.46	44.62	78.82	80.82	95.89	108.87	109.65	78.87
CQL2024.05	46.98	61.13	82.26	44.7	82.33	79.74	95.9	104.38	109.5	78.55
DW+CQL2024.10	46.5	66.1	82.1	45.1	88.6	75.3	86.1	92.9	109.7	692.4
DT2024.05	43.83	64.57	74.94	37.98	78.43	67.9	91.78	110.54	108.24	75.36
DT2024.05	42.44	64.49	72.22	38.71	66.47	59.67	91.8	110.57	108.4	72.75
DT2024.05	42.43	63.09	71.64	39.34	81.81	62.06	92.43	109.05	108.38	74.47
SDR2025.12	28.2	54	45.9	33.9	41.6	57	10.5	37.4	48.4	40.8
Random2025.12	26.3	50	41.9	28	32.8	47.8	10.1	32.7	32.2	27
GradMatch2025.12	6.8	18.3	13.6	3.3	29.3	23.9	18.9	43.7	44.5	22.5
Top Q-value2025.12	2.3	7.7	13.1	4.4	6.1	3.8	1.7	9.1	4.3	5.8
GraNd2025.12	2.1	14.5	5.9	2	6.4	2.7	3.1	1.3	3.3	4.6
Top Reward2025.12	1.4	6.9	2.7	10.6	5.9	3.6	2.2	3.9	7.6	5
Herding2025.12	1	4.4	2.7	0.7	0.6	3.6	3.3	2.9	1.1	2.3