Label Distribution Learning by Regularized Sample Self-Representation

Table 3

Chebyshev distance   (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	233.5 26.0	139.8 1.4	129.7 3.0	126.6 3.6	113.4 0.7	114.1 3.6
SBU-3DFE	142.2 5.4	144.2 6.1	133.2 4.7	104.2 4.5	124.0 4.9	107.6 2.4
SJAFFE	121.0 10.3	136.3 16.5	117.2 8.5	105.2 15.0	96.3 18.9	90.7 9.3
Yeast-alpha	13.8 0.4	37.6 2.4	16.9 0.3	13.4 0.4	13.4 0.4	13.4 0.5
Yeast-cdc	17.2 0.8	38.0 1.9	20.0 0.5	16.2 0.4	16.2 0.5	16.2 0.5
Yeast-cold	57.8 3.9	57.8 2.2	56.7 1.9	51.1 2.2	51.0 1.7	50.9 1.9
Yeast-diau	43.2 3.9	49.1 1.9	41.2 1.3	36.9 1.1	36.9 1.0	36.9 1.3
Yeast-dtt	38.6 2.2	44.6 2.5	43.3 1.6	36.0 1.8	35.9 1.5	35.9 1.7
Yeast-elu	17.0 0.3	37.8 2.5	20.2 0.8	16.3 0.5	16.2 0.5	16.2 0.5
Yeast-heat	44.0 1.0	55.1 4.0	46.5 1.1	42.3 1.0	42.2 1.5	42.2 1.1
Yeast-spo	64.7 3.0	66.3 2.9	61.7 1.7	58.3 2.8	58.2 2.7	58.0 2.2
Yeast-spo5	92.9 3.9	95.7 4.2	94.6 2.4	91.4 3.1	91.1 3.9	91.2 3.4