Enhancing Certified Robustness via Block Reflector Orthogonal Layers and Logit Annealing Loss
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915427870834688 |
|---|---|
| author | Lai, Bo-Han Huang, Pin-Han Kung, Bo-Han Chen, Shang-Tse |
| author_facet | Lai, Bo-Han Huang, Pin-Han Kung, Bo-Han Chen, Shang-Tse |
| contents | Lipschitz neural networks are well-known for providing certified robustness in deep learning. In this paper, we present a novel, efficient Block Reflector Orthogonal (BRO) layer that enhances the capability of orthogonal layers on constructing more expressive Lipschitz neural architectures. In addition, by theoretically analyzing the nature of Lipschitz neural networks, we introduce a new loss function that employs an annealing mechanism to increase margin for most data points. This enables Lipschitz models to provide better certified robustness. By employing our BRO layer and loss function, we design BRONet - a simple yet effective Lipschitz neural network that achieves state-of-the-art certified robustness. Extensive experiments and empirical analysis on CIFAR-10/100, Tiny-ImageNet, and ImageNet validate that our method outperforms existing baselines. The implementation is available at https://github.com/ntuaislab/BRONet. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_15174 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enhancing Certified Robustness via Block Reflector Orthogonal Layers and Logit Annealing Loss Lai, Bo-Han Huang, Pin-Han Kung, Bo-Han Chen, Shang-Tse Machine Learning Lipschitz neural networks are well-known for providing certified robustness in deep learning. In this paper, we present a novel, efficient Block Reflector Orthogonal (BRO) layer that enhances the capability of orthogonal layers on constructing more expressive Lipschitz neural architectures. In addition, by theoretically analyzing the nature of Lipschitz neural networks, we introduce a new loss function that employs an annealing mechanism to increase margin for most data points. This enables Lipschitz models to provide better certified robustness. By employing our BRO layer and loss function, we design BRONet - a simple yet effective Lipschitz neural network that achieves state-of-the-art certified robustness. Extensive experiments and empirical analysis on CIFAR-10/100, Tiny-ImageNet, and ImageNet validate that our method outperforms existing baselines. The implementation is available at https://github.com/ntuaislab/BRONet. |
| title | Enhancing Certified Robustness via Block Reflector Orthogonal Layers and Logit Annealing Loss |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2505.15174 |