🏢 Tübingen AI Center, University of Tübingen
Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness
·1858 words·9 mins·
loading
·
loading
AI Theory
Robustness
🏢 Tübingen AI Center, University of Tübingen
This paper optimizes randomized smoothing, a crucial certified defense against adversarial attacks, by introducing novel statistical methods that drastically reduce the computational cost, leading to …
FSP-Laplace: Function-Space Priors for the Laplace Approximation in Bayesian Deep Learning
·3503 words·17 mins·
loading
·
loading
AI Generated
Machine Learning
Deep Learning
🏢 Tübingen AI Center, University of Tübingen
FSP-LAPLACE efficiently integrates interpretable function-space priors into Bayesian deep learning via a novel Laplace approximation, significantly improving uncertainty estimates and model performanc…