# Provable Adversarial Robustness for Fractional Lp Threat Models

Source: [Provable Adversarial Robustness for Fractional Lp Threat Models](<https://www.sota2.com/research/paper/68fa6bfd1e209131cea31749>)

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

Data updated: 2025-10-26T00:01:55.246Z

Authors: Alexander Levine, Soheil Feizi

Published: 2022-03-16T21:11:41.000Z

## Abstract

In recent years, researchers have extensively studied adversarial robustness in a variety of threat models, including L\_0, L\_1, L\_2, and L\_infinity-norm bounded adversarial attacks. However, attacks bounded by fractional L\_p "norms" (quasi-norms defined by the L\_p distance with 0\<p\<1) have yet to be thoroughly considered. We proactively propose a defense with several desirable properties: it provides provable (certified) robustness, scales to ImageNet, and yields deterministic (rather than high-probability) certified guarantees when applied to quantized data (e.g., images). Our technique for fractional L\_p robustness constructs expressive, deep classifiers that are globally Lipschitz with respect to the L\_p^p metric, for any 0\<p\<1. However, our method is even more general: we can construct classifiers which are globally Lipschitz with respect to any metric defined as the sum of concave functions of components. Our approach builds on a recent work, Levine and Feizi (2021), which provides a provable defense against L\_1 attacks. However, we demonstrate that our proposed guarantees are highly non-vacuous, compared to the trivial solution of using (Levine and Feizi, 2021) directly and applying norm inequalities. Code is available at https://github.com/alevine0/fractionalLpRobustness.

## 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/2203.08945>)

[PDF](<https://arxiv.org/pdf/2203.08945.pdf>)
