AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Bayesian model selection and misspecification testing in imaging inverse problems only from noisy and partial measurements

Tom Sprunck, Marcelo Pereyra, Tobías I. Liaudat

ICML 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Simulation-based inference · 0 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
use Phi1 data-fission estimator instead of Bayesian residualblur kernel selection · single shot accuracyaccuracy: 40.0% -> 86.7%Improved
use Phi1 instead of EB residualblur kernel selection · single shot accuracyaccuracy: 46.7% -> 86.7%Improved
pool three measurements (few shot) instead of single shotPhi1 · blur kernel selectionaccuracy: 86.7% -> 100%Improved
increase blur strength sigma from 0.5 to 5Phi2 OOD test · moderate OOD (Met) deblurringpower: 87% -> 60%Hurt
increase blur strength sigma from 0.5 to 5Phi2 OOD test · strong OOD deblurringpower: 100% -> 100%No clear effect
use Phi2 instead of Phi1 for OOD detectionMRI R=4 · brain vs knee powerpower: 94% -> 100%Improved
use Phi2 instead of Phi1 for OOD detectionMRI R=8 · brain vs knee powerpower: 96% -> 100%Improved
decrease splitting parameter alphaPhi estimator · Gaussian toy modelmodel discriminationImproved
average K noise realizationsestimator · Gaussian toy modelbiasImproved
decrease alphaestimator · Gaussian toy modelcomputational costHurt
use DRUNet prior instead of wavelet priorPhi1 · galaxy denoisingPhi1: 83.38 -> 22.23Improved
use DRUNet prior instead of DnCNNposterior mean PSNR · galaxy denoisingPSNR (dB): 29.86 -> 32.86Improved

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