AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Solving Inverse Problems With Deep Neural Networks – Robustness Included?

Martin Genzel, Jan Macdonald, Maximilian März

TPAMI 2022 · Adversarial robustness and security › Certified and empirical robustness · approach: Decision-boundary geometry of robustness · 165 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
train with noise jittering instead of noiseless data (ablation)ItNet · MNIST compressed sensing (Scenario A2), adversarial noiserelative reconstruction error: error blows up at ~15% adversarial noise without jitteringImproved
train with noise jittering instead of noiseless data (ablation)UNet · Radon ellipses (Scenario B2), 1% adversarial noisereconstruction robustnessImproved
use original fastMRI vertical-line mask instead of radial maskTira and TV · fastMRI single-coil, 4-fold accelerationadversarial robustness: undersampling ~23% vs ~17%Hurt
use ItNet data-consistent iterative architecture instead of plain post-processingItNet · CS Gaussian measurementsrobustness and accuracyImproved
use Tiramisu instead of U-NetTiramisu · CS Gaussian measurementsreconstruction accuracyImproved
make inversion layer learnable (fully-learned) (ablation)UNetFL/ItNetFL · CS Gaussian measurementsreconstruction accuracyImproved
fully-learned inversion layer instead of Fourier adjointfully learned schemes · Fourier ellipses (Scenario B1)reconstruction accuracyNo clear effect
replace TV minimization with learned networks under statistical noiseNN-based methods · all case studies, mid to high noiserelative errorImproved
replace TV minimization with learned networks under adversarial noiseNN-based methods · all case studiesrelative errorNo clear effect
add a ConvNet classifier after reconstruction (compressed classification)Rec + ConvNet · MNIST from Gaussian measurementsaccuracy: drops to zero at some noise levelHurt

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