RoBIC: A benchmark suite for assessing classifiers robustness - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year :

RoBIC: A benchmark suite for assessing classifiers robustness


Many defenses have emerged with the development of adversarial attacks. Models must be objectively evaluated accordingly. This paper systematically tackles this concern by proposing a new parameter-free benchmark we coin RoBIC. RoBIC fairly evaluates the robustness of image classifiers using a new half-distortion measure. It gauges the robustness of the network against white and black box attacks, independently of its accuracy. RoBIC is faster than the other available benchmarks. We present the significant differences in the robustness of 16 recent models as assessed by RoBIC. We make this benchmark publicly available for use and contribution at robustness_benchmark.
Fichier principal
Vignette du fichier
main.pdf (258.72 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03234791 , version 1 (25-05-2021)



Thibault Maho, Benoît Bonnet, Teddy Furon, Erwan Le Merrer. RoBIC: A benchmark suite for assessing classifiers robustness. ICIP 2021 - IEEE International Conference on Image Processing, Sep 2021, Anchorage, Alaska, United States. pp.1-5, ⟨10.1109/ICIP42928.2021.9506053⟩. ⟨hal-03234791⟩
70 View
101 Download



Gmail Facebook Twitter LinkedIn More