TY  - JOUR
AU  - Strisciuglio, Nicola
AU  - Lopez-Antequera, Manuel
AU  - Petkov, Nicolai
PY  - 2020
DO  - 10.1007/s00521-020-04751-8
SN  - 0941-0643
UR  - http://hdl.handle.net/10668/18637
T2  - Neural computing & applications
AB  - Convolutional neural networks (CNNs) lack robustness to test image corruptions that are not seen during training. In this paper, we propose a new layer for CNNs that increases their robustness to several types of corruptions of the input images. We...
LA  - en
PB  - Springer london ltd
KW  - Convolutional neural networks
KW  - Image corruption
KW  - Network robustness
KW  - Neuron response inhibition
KW  - push-pull layer
KW  - Orientation selectivity
KW  - Receptive-fields
KW  - Simple cell
KW  - Suppression
KW  - Broad
KW  - Model
KW  - Benchmarking
KW  - Neural Networks, Computer
KW  - Inhibition, Psychological
KW  - Algorithms
TI  - Enhanced robustness of convolutional networks with a push-pull inhibition layer
TY  - research article
VL  - 32
ER  -