AI Paper Atlas

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

Structuring Uncertainty for Fine-Grained Sampling in Stochastic Segmentation Networks

Frank Nussbaum, Jakob Gawlikowski, Julia Niebling

NeurIPS 2022 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Priors and functional variational uncertainty · 2 citations (OpenAlex)

Read the paper

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 flow-probability (FP) rotations instead of classic rotationsSSN factors · LIDCrow sparsity: 48.2 -> 82.8Improved
use FP rotations instead of classic rotationsSSN factors · SEN12MSrow sparsity: 52.7 -> 74.1Improved
use FP rotations instead of classic rotationsSSN factors · CamVidrow sparsity: 30.4 -> 50.1Improved
apply classic rotation (Varimax) to unrotated factorsSSN factors · LIDCrow sparsity: 29.3 -> 48.2Improved
use FP-Quartimax instead of FP-VarimaxSSN factors · SEN12MSrow sparsity: 74.1 -> 77.7Improved
use FP-Quartimax instead of FP-VarimaxSSN factors · LIDCrow sparsity: 82.8 -> 82.7No clear effect
use FP rotations instead of classic rotationsSSN factors · CamVidnumber of relevant factorsImproved
use classic rotations instead of unrotated factorsSSN factors · LIDC, SEN12MS, CamVidnumber of relevant factorsNo clear effect
use FP rotations instead of classic rotationsSSN factors · LIDC, SEN12MS, CamVidAUC of factor separationImproved
use classic rotations instead of unrotated factorsSSN factors · LIDC, SEN12MS, CamVidAUC of factor separationImproved

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