TY  - JOUR
AU  - Ivet Rafegas
AU  - Maria Vanrell
AU  - Luis A Alexandre
AU  - G. Arias
PY  - 2020//
TI  - Understanding trained CNNs by indexing neuron selectivity
T2  - PRL
JO  - Pattern Recognition Letters
SP  - 318
EP  - 325
VL  - 136
N2  - The impressive performance of Convolutional Neural Networks (CNNs) when solving different vision problems is shadowed by their black-box nature and our consequent lack of understanding of the representations they build and how these representations are organized. To help understanding these issues, we propose to describe the activity of individual neurons by their Neuron Feature visualization and quantify their inherent selectivity with two specific properties. We explore selectivity indexes for: an image feature (color); and an image label (class membership). Our contribution is a framework to seek or classify neurons by indexing on these selectivity properties. It helps to find color selective neurons, such as a red-mushroom neuron in layer Conv4 or class selective neurons such as dog-face neurons in layer Conv5 in VGG-M, and establishes a methodology to derive other selectivity properties. Indexing on neuron selectivity can statistically draw how features and classes are represented through layers in a moment when the size of trained nets is growing and automatic tools to index neurons can be helpful.
UR  - https://doi.org/10.1016/j.patrec.2019.10.013
L1  - http://refbase.cvc.uab.es/files/RVL2019.pdf
N1  - CIC; 600.087; 600.140; 600.118
ID  - Ivet Rafegas2020
ER  -