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

Continual Reinforcement Learning on DMC6

1.04Normalized Score

FT-MT-pre-trained

Updated 1mo ago

Evaluation Results

Method	Links
FT-MT-pre-trained 2023.06		1.04	764.38	-
FT-MT-pre-trained 2023.06		1.04	764.38	-
FT-MT-scratch 2023.06		0.92	648.49	-
FT-MT-scratch 2023.06		0.92	648.49	-
L2M-oracle 2023.06		0.7	549.29	-0.02
L2M-oracle 2023.06		0.7	549.29	-0.02
L2M 2023.06		0.56	401.65	0.11
L2M 2023.06		0.56	401.65	0.11
L2P-Pv2 2023.06		0.32	271.39	0.02
L2P-Pv2 2023.06		0.32	271.39	0.02
EWC 2023.06		0.3	244.7	0.49
EWC 2023.06		0.3	244.7	0.49
FT 2023.06		0.27	225.81	0.83
FT 2023.06		0.27	225.81	0.83
FT-last+head 2023.06		0.25	204.49	0.51
FT-last+head 2023.06		0.25	204.49	0.51
L2P-PreT 2023.06		0.2	182.33	-0.03
L2 2023.06		0.2	192.8	-0.01
L2P-PreT 2023.06		0.2	182.33	-0.03
L2 2023.06		0.2	192.8	-0.01
L2P-PT 2023.06		0.18	158.09	-0.06
L2P-PT 2023.06		0.18	158.09	-0.06
FT-head 2023.06		0.06	58.84	-0.02
FT-head 2023.06		0.06	58.84	-0.02