Label Distribution Learning by Regularized Sample Self-Representation

Table 5

Canberra Meric   (mean std) × 10³ of different algorithms on the twelve datasets. The best results are enlightened in bold and the second best results are italicized.

Algorithms	PT-SVM	AA-BP	SA-IIS	SA-BFGS	RSSR-LDL2	RSSR-LDL21
Movie	1693 183.1	1232 22.2	1063 27.0	1063 22.9	992.0 10.3	989.1 24.6
SBU-3DFE	925.7 34.3	984.1 47.6	888.8 20.6	725.1 24.9	836.2 14.3	782.5 8.6
SJAFFE	917.8 59.2	1034.8 97.7	870.6 44.4	862.5 76.1	735.9 80.7	705.8 46.8
Yeast-alpha	723.1 19.3	2483.5 172.3	859.2 16.0	681.9 21.1	679.0 16.7	678.3 19.1
Yeast-cdc	685.7 24.7	1772.5 74.3	786.7 13.4	647.3 14.9	645.0 14.6	642.3 14.9
Yeast-cold	269.5 17.6	269.9 8.6	264.5 10.0	240.1 10.0	239.7 8.6	239.4 9.7
Yeast-diau	508.8 47.2	584.7 28.5	480.8 13.9	430.5 15.4	429.9 10.6	429.7 11.2
Yeast-dtt	179.9 10.1	209.4 11.3	201.0 8.8	169.0 8.8	168.6 5.9	168.4 8.4
Yeast-elu	621.2 16.2	1546.8 120.2	714.7 18.0	582.6 18.0	581.1 11.5	581.0 13.1
Yeast-heat	380.7 11.8	486.5 40.4	403.3 9.9	364.4 9.0	363.5 12.0	362.8 7.6
Yeast-spo	562.8 21.8	589.2 24.6	541.6 12.8	512.9 24.1	512.5 23.4	511.8 18.3
Yeast-spo5	287.3 12.8	295.5 13.3	292.3 7.3	283.1 10.5	282.1 12.9	282.3 10.5