Batch Reinforcement Learning on D4RL v0 (test) benchmark leaderboard · SOTA2 Research

Batch Reinforcement Learning on D4RL v0 (test)

32.8HalfCheetah Score (Random)CDC

Updated 5mo ago

Evaluation Results

Method	Links
CDC2021.02		32.8	49.51	22.72	7.12	-0.95	1.58	0.58	16.4	18.07	1.27	2.96	0.33	3.81	2.65	-0.1	22.22	0	0	0	0	0	-3.43	-3.4	0.26	0.26	-0.16	-0.36	-0.31	-0.15	0	0	0	173.67	0
SAC2021.02		29.6	40.97	26.47	3.78	-0.41	0.8	0.81	0.59	8.96	0.8	1.3	0.81	0.04	4.09	0.05	0	0	0	0	0	0	-1.15	-0.64	0.26	0.27	-0.34	-0.34	-0.31	-0.11	0	0	0	116.28	0
BRAC-V2021.02		28.1	45.5	45.9	45.3	-1.1	12	32.3	0.8	0.8	3.7	0.5	81.3	0.9	0.9	0	70	70	0	0	0	0	0.6	-2.5	0.2	0.3	-0.3	-0.1	-0.3	-0.3	0	0	0	434.5	0
CDC2021.02		27.36	46.05	44.74	59.64	82.05	14.76	60.39	55.89	86.9	102.75	7.22	82.13	22.96	70.91	87.54	91.85	62.59	55.19	40.74	5.19	11.85	73.19	49.18	4.34	2.37	4.62	0.01	0.73	-0.24	58.7	42.5	42.87	1,396.99	—
BEAR2021.02		25.5	38.6	36.2	51.7	108.2	9.5	47.6	25.3	4	110.3	6.7	33.2	10.8	10.8	106.1	73	61	0	0	0	0	-1	26.5	0.3	0.3	-0.3	-0.1	-0.3	-0.2	0	13.1	47.2	844	0.0001
BRAC-P2021.02		23.5	44	45.6	43.8	3.8	11.1	31.2	0.7	1.1	6.6	0.8	72.7	-0.3	-0.3	-0.2	50	40	0	0	0	0	8.1	1.6	0.3	0.3	-0.3	-0.1	-0.3	-0.3	0	0	0	383.4	0
CQL2021.02		16.71	38.97	42.77	39.18	85.49	10.37	51.79	28.67	112.46	109.97	2.77	71.03	19.95	90.55	103.6	90.12	11.11	40.74	40.74	0	0	37.5	39.2	4.4	2.1	9.9	0.4	0.2	-0.1	43.8	49.8	51	1,245.2	0.036
BCQ2021.02		2.25	41.48	34.79	69.64	97.44	10.6	65.07	31.05	109.1	75.52	4.31	52.03	13.67	67.26	87.59	88.52	61.11	0	0	1.85	0	66.88	50.86	0.91	0.38	-0.05	0.01	-0.04	-0.28	0.83	9.26	18.43	1,060.46	0.0006
BC2021.02		2.1	36.1	38.4	35.8	107	9.8	29	11.8	119.9	109	1.6	6.6	11.3	11.3	125.7	65	55	0	0	0	0	34.4	56.9	1.5	0.8	0.5	-0.1	0	-0.1	33.8	33.8	47.5	984.4	0.0016