← Search

Camila Gonzalez

2 accepted papers

2026

Does a Hybrid Space-Aware Randomized Defense Improve Empirical and Certified Adversarial Robustness?

ICML 2026poster

We introduce Hybrid Space-aware Stochastic Convolution Attention Noise (HySCAN), a hybrid randomized defense that helps close the long-standing gap between provable robustness under ℓ2 certificates and empirical robustness against strong ℓ∞ attacks, while maintaining strong generalization across div…

Cited by 0SourceScholar
2025

Confounder-Free Continual Learning via Recursive Feature Normalization

ICML 2025poster

Confounders are extraneous variables that affect both the input and the target, resulting in spurious correlations and biased predictions. There are recent advances in dealing with or removing confounders in traditional models, such as metadata normalization (MDN), where the distribution of the lear…

Cited by 0SourcePDFScholar