Back to Search
Start Over
Defense against adversarial attacks on deep convolutional neural networks through nonlocal denoising
- Source :
- IAES International Journal of Artificial Intelligence (IJ-AI). 11:961
- Publication Year :
- 2022
- Publisher :
- Institute of Advanced Engineering and Science, 2022.
-
Abstract
- Despite substantial advances in network architecture performance, the susceptibility of adversarial attacks makes deep learning challenging to implement in safety-critical applications. This paper proposes a data-centric approach to addressing this problem. A nonlocal denoising method with different luminance values has been used to generate adversarial examples from the Modified National Institute of Standards and Technology database (MNIST) and Canadian Institute for Advanced Research (CIFAR-10) data sets. Under perturbation, the method provided absolute accuracy improvements of up to 9.3% in the MNIST data set and 13% in the CIFAR-10 data set. Training using transformed images with higher luminance values increases the robustness of the classifier. We have shown that transfer learning is disadvantageous for adversarial machine learning. The results indicate that simple adversarial examples can improve resilience and make deep learning easier to apply in various applications.
- Subjects :
- FOS: Computer and information sciences
Denoising
Computer Science - Machine Learning
Computer Science - Cryptography and Security
Information Systems and Management
Computer Vision and Pattern Recognition (cs.CV)
Adversarial attacks
Computer Science - Computer Vision and Pattern Recognition
Deep learning
Adversarial machine learning
Machine Learning (cs.LG)
Artificial Intelligence
Control and Systems Engineering
Convolutional neural networks
Electrical and Electronic Engineering
Cryptography and Security (cs.CR)
Subjects
Details
- ISSN :
- 22528938 and 20894872
- Volume :
- 11
- Database :
- OpenAIRE
- Journal :
- IAES International Journal of Artificial Intelligence (IJ-AI)
- Accession number :
- edsair.doi.dedup.....b85149f67d728ef58927ff505269f5e3