Szegedy, Vanhoucke, Ioffe, Shlens, Wojna - «Rethinking the Inception Architecture for Computer Vision» (2015)

Theory
, , ,
2

Some citations

GoogLeNet employed only 5 million parameters, which represented a 12x reduction with respect to its predecessor AlexNet, which used 60 million parameters.
Furthermore, VGGNet employed about 3x more parameters than AlexNet.