Image Classification on CIFAR-10 OOD: SVHN (test) benchmark leaderboard · SOTA2 Research

Image Classification on CIFAR-10 OOD: SVHN (test)

0.843AUROC (H)Deep ensemble

Evaluation Results

Method	Links
Deep ensemble2021.06		0.843	0.736	85.552	2.244	1.667	0.049	0.277
kde-WGD2021.06		0.838	0.735	85.904	2.205	1.661	0.053	0.276
sge-WGD2021.06		0.837	0.725	85.792	2.214	1.634	0.051	0.275
ssge-WGD2021.06		0.832	0.731	85.638	2.182	1.655	0.049	0.276
SVGD2021.06		0.825	0.71	85.142	2.106	1.567	0.052	0.287
sge-fWGD2021.06		0.795	0.754	84.766	1.984	1.729	0.047	0.288
ssge-fWGD2021.06		0.792	0.752	84.762	1.97	1.723	0.046	0.286
kde-fWGD2021.06		0.791	0.758	84.888	1.97	1.749	0.044	0.282
hyper-DE2021.06		0.789	0.743	84.743	1.951	1.69	0.046	0.288
f-SVGD2021.06		0.783	0.712	84.51	1.968	1.624	0.049	0.292