Gaussian class-conditional simplex loss for accurate, adversarially robust deep classifier training

Gaussian class-conditional simplex loss for accurate, adversarially robust deep classifier training

Abstract In this work, we present the Gaussian Class-Conditional Simplex (GCCS) loss: a novel approach for training deep robust multiclass classifiers that improves over the state-of-the-art in terms of classification accuracy and adversarial robustness, with little extra cost for network training....

Full description

Bibliographic Details
Main Authors: Arslan Ali, Andrea Migliorati, Tiziano Bianchi, Enrico Magli
Format: Article
Language:English
Published: SpringerOpen 2023-03-01
Series:EURASIP Journal on Information Security
Subjects:
Deep neural networks
Classification
Adversarial robustness
Adversarial training
Adversarial defense
Online Access:https://doi.org/10.1186/s13635-023-00137-0

Internet

https://doi.org/10.1186/s13635-023-00137-0

Similar Items