Anomaly Detection on MSDS benchmark leaderboard · SOTA2 Research

Anomaly Detection on MSDS

0.9105AUCGDN

Evaluation Results

Method	Links
GDN2022.01		0.9105	0.9989	0.8026	0.89
DAGMM2022.01		0.9013	0.9891	0.8026	0.8861
CAE-M2022.01		0.9013	0.9908	0.8439	0.9115
TranAD2022.01		0.9013	0.9999	0.8626	0.9262
OmniAnomaly2022.01		0.8982	1	0.7964	0.8867
MTAD-GAT2022.01		0.8982	0.9919	0.7964	0.8835
USAD2022.01		0.8979	0.9912	0.7959	0.8829
MSCRED2022.01		0.8943	1	0.7983	0.8878
TranAD2022.01		0.8113	—	—	0.8391
MAD-GAN2022.01		0.8054	0.9982	0.6107	0.7578
LSTM-NDT2022.01		0.8013	0.9999	0.8012	0.8896
LSTM-NDT2022.01		0.7813	—	—	0.7912
DAGMM2022.01		0.7763	—	—	0.8389
MSCRED2022.01		0.7716	—	—	0.8283
USAD2022.01		0.7613	—	—	0.8389
GDN2022.01		0.6819	—	—	0.8389
MTAD-GAT2022.01		0.6122	—	—	0.8248
CAE-M2022.01		0.6001	—	—	0.8389
OmniAnomaly2022.01		0.5613	—	—	0.8389
MERLIN2022.01		0.5022	0.7254	0.311	0.4353
MERLIN2022.01		0.5022	—	—	0.4353
MAD-GAN2022.01		0.5002	—	—	0.739