AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Semantic Adversarial Attacks: Parametric Transformations That Fool Deep Classifiers

Ameya Joshi, Amitangshu Mukherjee, Soumik Sarkar, Chinmay Hegde

ICCV 2019 · Adversarial robustness and security › Attacks on text and graphs · approach: Adversarial perturbations beyond images · 96 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add attributes A5 and A6 to single-attribute attack A1 (multi-attribute Fader Network) (ablation)Adversarial Fader Network · CelebA gender classifier, 500 test imagestarget model accuracy (%): 52.0 -> 3.00Improved
add eyebrows attribute A9 to 5-attribute AttGAN attack (ablation)Adversarial AttGAN · CelebA gender classifier, 500 test imagestarget model accuracy (%): 70.40 -> 39.40Improved
optimize attribute parameters instead of worst-of-10 random sampling (single attribute A1)Adversarial Fader Network · CelebA gender classifier, 500 test imagestarget model accuracy (%): 89.00 -> 52.0Improved
optimize attribute parameters instead of random sampling (AttGAN, 5 attributes)Adversarial AttGAN · CelebA gender classifier, 500 test imagestarget model accuracy (%): 32.80 -> 70.40Hurt
optimize attribute parameters instead of random sampling (AttGAN, 6 attributes)Adversarial AttGAN · CelebA gender classifier, 500 test imagestarget model accuracy (%): 40.40 -> 39.40No clear effect
multi-attribute semantic attack versus PGD at equal l_inf boundMulti-attribute semantic attack · CelebA gender classifier, pixel-space comparison (eps=1.74)target model accuracy (%): 26.2 -> 1.00Improved
multi-attribute semantic attack versus FGSMMulti-attribute semantic attack · CelebA gender classifier, pixel-space comparison (eps=1.74)target model accuracy (%): 91.6 -> 1.00Improved
multi-attribute semantic attack versus Spatial attackMulti-attribute semantic attack · CelebA gender classifier, pixel-space comparison (eps=1.74)target model accuracy (%): 41.00 -> 1.00Improved
multi-attribute semantic attack versus CW-l_infMulti-attribute semantic attack · CelebA gender classifier, pixel-space comparison (eps=1.74)target model accuracy (%): 1.00 -> 0.00Hurt
add attributes A5 and A6 to single-attribute attack A1 (ablation)Adversarial Fader Network · CelebA age classifier (appendix)target model accuracy (%): 70.0 -> 12.0Improved
increase dimensionality k of the semantic attack space (ablation)Synthetic subspace and neural attacks · Synthetic MoG binary classifiertarget model accuracy (%)Improved
add ReLU nonlinearity to the transformation function (ablation)Synthetic subspace and neural attacks · Synthetic MoG binary classifiertarget model accuracy (%)Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy