Kernel Negative ε Dragging Linear Regression for Pattern Classification

Table 2

Accuracies (%) of different methods on the FERET database.

Number of training samples per class	1	2	3	4	5
Our method	35.52	55.84	66.29	79.03	78.18
DLSR	19.45	31.18	36.99	45.95	44.15
CLSR	28.03	46.38	55.18	65.03	65.07
NNLS	34.26	53.48	64.38	75.84	76.33
-SVM	32.01	54.73	64.43	75.89	73.96
KNN	32.01	44.87	54.44	68.14	65.70
SRC	23.54	41.48	53.12	68.51	70.46
LRC	32.01	55.34	65.89	77.12	74.76
NDLR	33.77	54.28	63.50	72.64	72.60