AI Paper Atlas

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

Benchmarking Bayesian Deep Learning on Diabetic Retinopathy Detection Tasks

Neil Band, Tim G. J. Rudner, Qixuan Feng, Angelos Filos, Zachary Nado, Michael W Dusenberry, Ghassen Jerfel, Dustin Tran, Yarin Gal

NeurIPS 2021 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Single-model predictive uncertainty · 14 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 MC Dropout uncertainty for selective referral instead of MAP uncertaintyRetina benchmark models · Severity Shift: severe/proliferativereferral rate at 100% accuracy: nearly 20% lowerImproved
Use uncertainty-based referral as referral rate increasesmost methods · Country Shift: APTOSAUCHurt
Use uncertainty-based referral with MFVIMFVI · Country Shift: APTOSAUC: 91.4% up to 93.8%Mixed
Add epistemic uncertainty using MC Dropout rather than deterministic MAPRetina models · Severity Shift: severe/proliferativereferral rate needed for 100% accuracy: nearly 20% lowerImproved
Use variational inference methods, including MC Dropout, FSVI, and Deep Ensemble, instead of MAPRetina benchmark models · Country Shift: in-domainAUCImproved
Tune hyperparameters on balanced joint in-domain and shifted validation data instead of in-domain validation AUC (ablation)Retina benchmark models · Country Shift: APTOSaccuracy referral curveImproved
Use a single Monte Carlo sample at training time, enabling larger batch sizes (ablation)variational methods · Retina benchmark validationvalidation performanceImproved
Use constant class weights instead of mini-batch-relative class weighting (ablation)Retina benchmark training · diabetic retinopathy detectionoverall performanceHurt

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