AI Paper Atlas

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

Learning Provably Robust Estimators for Inverse Problems via Jittering

Anselm Krainovic, Mahdi Soltanolkotabi, Reinhard Heckel

NeurIPS 2023 · Adversarial robustness and security › Certified and empirical robustness · approach: Decision-boundary geometry of robustness · 2 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 the denoiser with Gaussian jittering instead of standard trainingU-net · Gaussian image denoisingworst-case riskImproved
Use jittering instead of adversarial trainingU-net · Gaussian image denoisingworst-case riskNo clear effect
Train with Gaussian jittering instead of standard trainingU-net · Image deconvolutionworst-case riskImproved
Use jittering instead of adversarial trainingU-net · Image deconvolutionworst-case riskHurt
Train with Gaussian jittering instead of standard trainingU-net · Compressive sensingworst-case riskImproved
Use jittering instead of adversarial trainingU-net · Compressive sensingworst-case riskHurt
Add Gaussian jittering noise during trainingLinear estimator · Subspace Gaussian denoisingworst-case riskNo clear effect
Increase the desired robustness levelOptimal robust denoiser · Subspace Gaussian denoisingstandard riskHurt
Optimize the jittering noise level for the desired robustnessU-net · Gaussian image denoisingworst-case riskImproved
Use jittering rather than adversarial trainingU-net · Inverse problemstraining computational costImproved
Use jittering for a general linear inverse problem with unequal singular valuesLinear estimator · General linear inverse problemsworst-case riskHurt

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