Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2412.19747 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915081504161792 |
|---|---|
| author | Wang, Longwei Nayyem, Navid Rakin, Abdullah |
| author_facet | Wang, Longwei Nayyem, Navid Rakin, Abdullah |
| contents | Adversarial attacks exploit the vulnerabilities of convolutional neural networks by introducing imperceptible perturbations that lead to misclassifications, exposing weaknesses in feature representations and decision boundaries. This paper presents a novel framework combining supervised contrastive learning and margin-based contrastive loss to enhance adversarial robustness. Supervised contrastive learning improves the structure of the feature space by clustering embeddings of samples within the same class and separating those from different classes. Margin-based contrastive loss, inspired by support vector machines, enforces explicit constraints to create robust decision boundaries with well-defined margins. Experiments on the CIFAR-100 dataset with a ResNet-18 backbone demonstrate robustness performance improvements in adversarial accuracy under Fast Gradient Sign Method attacks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_19747 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing Adversarial Robustness of Deep Neural Networks Through Supervised Contrastive Learning Wang, Longwei Nayyem, Navid Rakin, Abdullah Machine Learning Artificial Intelligence Adversarial attacks exploit the vulnerabilities of convolutional neural networks by introducing imperceptible perturbations that lead to misclassifications, exposing weaknesses in feature representations and decision boundaries. This paper presents a novel framework combining supervised contrastive learning and margin-based contrastive loss to enhance adversarial robustness. Supervised contrastive learning improves the structure of the feature space by clustering embeddings of samples within the same class and separating those from different classes. Margin-based contrastive loss, inspired by support vector machines, enforces explicit constraints to create robust decision boundaries with well-defined margins. Experiments on the CIFAR-100 dataset with a ResNet-18 backbone demonstrate robustness performance improvements in adversarial accuracy under Fast Gradient Sign Method attacks. |
| title | Enhancing Adversarial Robustness of Deep Neural Networks Through Supervised Contrastive Learning |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2412.19747 |