# Robust Principles: Architectural Design Principles for Adversarially Robust CNNs Source: [Robust Principles: Architectural Design Principles for Adversarially Robust CNNs]() Collection index: [llms.txt]() Data updated: 2025-10-26T01:42:00.795Z Authors: ShengYun Peng, Weilin Xu, Cory Cornelius, Matthew Hull, Kevin Li, Rahul Duggal, Mansi Phute, Jason Martin, Duen Horng Chau Published: 2023-08-30T18:31:51.000Z ## Abstract Our research aims to unify existing works' diverging opinions on how architectural components affect the adversarial robustness of CNNs. To accomplish our goal, we synthesize a suite of three generalizable robust architectural design principles: (a) optimal range for depth and width configurations, (b) preferring convolutional over patchify stem stage, and (c) robust residual block design through adopting squeeze and excitation blocks and non-parametric smooth activation functions. Through extensive experiments across a wide spectrum of dataset scales, adversarial training methods, model parameters, and network design spaces, our principles consistently and markedly improve AutoAttack accuracy: 1-3 percentage points (pp) on CIFAR-10 and CIFAR-100, and 4-9 pp on ImageNet. The code is publicly available at https://github.com/poloclub/robust-principles. ## 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. - [Adversarial Robustness on CIFAR-10 (test)](): Adversarial Robustness; CIFAR-10 (test); Attack Success Rate (ASR): not recorded. ## Original sources and implementation [Original paper]() [PDF]()