Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909867039522816 |
|---|---|
| author | Ribeiro, Antônio H. Vävinggren, David Zachariah, Dave Schön, Thomas B. Bach, Francis |
| author_facet | Ribeiro, Antônio H. Vävinggren, David Zachariah, Dave Schön, Thomas B. Bach, Francis |
| contents | Adversarial training has emerged as a key technique to enhance model robustness against adversarial input perturbations. Many of the existing methods rely on computationally expensive min-max problems that limit their application in practice. We propose a novel formulation of adversarial training in reproducing kernel Hilbert spaces, shifting from input to feature-space perturbations. This reformulation enables the exact solution of inner maximization and efficient optimization. It also provides a regularized estimator that naturally adapts to the noise level and the smoothness of the underlying function. We establish conditions under which the feature-perturbed formulation is a relaxation of the original problem and propose an efficient optimization algorithm based on iterative kernel ridge regression. We provide generalization bounds that help to understand the properties of the method. We also extend the formulation to multiple kernel learning. Empirical evaluation shows good performance in both clean and adversarial settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_20883 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization Ribeiro, Antônio H. Vävinggren, David Zachariah, Dave Schön, Thomas B. Bach, Francis Machine Learning Cryptography and Security Optimization and Control Adversarial training has emerged as a key technique to enhance model robustness against adversarial input perturbations. Many of the existing methods rely on computationally expensive min-max problems that limit their application in practice. We propose a novel formulation of adversarial training in reproducing kernel Hilbert spaces, shifting from input to feature-space perturbations. This reformulation enables the exact solution of inner maximization and efficient optimization. It also provides a regularized estimator that naturally adapts to the noise level and the smoothness of the underlying function. We establish conditions under which the feature-perturbed formulation is a relaxation of the original problem and propose an efficient optimization algorithm based on iterative kernel ridge regression. We provide generalization bounds that help to understand the properties of the method. We also extend the formulation to multiple kernel learning. Empirical evaluation shows good performance in both clean and adversarial settings. |
| title | Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization |
| topic | Machine Learning Cryptography and Security Optimization and Control |
| url | https://arxiv.org/abs/2510.20883 |