# $\\sigma$-zero: Gradient-based Optimization of $\\ell\_0$-norm Adversarial Examples

Source: [$\\sigma$-zero: Gradient-based Optimization of $\\ell\_0$-norm Adversarial Examples](<https://www.sota2.com/research/paper/68fa774a1e209131cea72510>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T04:14:52.638Z

Authors: Antonio Emanuele Cin\\\`a, Francesco Villani, Maura Pintor, Lea Sch\\"onherr, Battista Biggio, Marcello Pelillo

Published: 2024-02-02T20:08:11.000Z

## Abstract

Evaluating the adversarial robustness of deep networks to gradient-based attacks is challenging. While most attacks consider $\\ell\_2$- and $\\ell\_\\infty$-norm constraints to craft input perturbations, only a few investigate sparse $\\ell\_1$- and $\\ell\_0$-norm attacks. In particular, $\\ell\_0$-norm attacks remain the least studied due to the inherent complexity of optimizing over a non-convex and non-differentiable constraint. However, evaluating adversarial robustness under these attacks could reveal weaknesses otherwise left untested with more conventional $\\ell\_2$- and $\\ell\_\\infty$-norm attacks. In this work, we propose a novel $\\ell\_0$-norm attack, called $\\sigma$-zero, which leverages a differentiable approximation of the $\\ell\_0$ norm to facilitate gradient-based optimization, and an adaptive projection operator to dynamically adjust the trade-off between loss minimization and perturbation sparsity. Extensive evaluations using MNIST, CIFAR10, and ImageNet datasets, involving robust and non-robust models, show that $\\sigma$\\texttt{-zero} finds minimum $\\ell\_0$-norm adversarial examples without requiring any time-consuming hyperparameter tuning, and that it outperforms all competing sparse attacks in terms of success rate, perturbation size, and efficiency.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2402.01879>)

[PDF](<https://arxiv.org/pdf/2402.01879.pdf>)
