Improving the Performance of Deep Learning Model-Based Classification by the Analysis of Local Probability

Table 2

Result on CIFAR-10.

Methods	Zero20 (%)	Zero40 (%)	Zero80 (%)
VoVNet-57 [24]	82.13	81.41	82.185
VGG16 [25]	72.95	70.39	71.39
ResNeSt50 [26]	69.17	69.94	68.84
L-PDL-joint (our)	84.69	87.24	95.24
L-PDL-weight (our)	84.68	87.21	95.17