AI Paper Atlas

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

Specifying Weight Priors in Bayesian Deep Neural Networks with Empirical Bayes

Ranganath Krishnan, Mahesh Subedar, Omesh Tickoo

AAAI 2020 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Variational Bayesian deep learning · 44 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
Set Gaussian weight-prior means and variational initialization from pretrained deterministic DNN weightsMFVI with MOPED · UCF-101validation accuracy: 0.029 -> 0.867Improved
Set Gaussian weight-prior means and variational initialization from pretrained deterministic DNN weightsMFVI with MOPED · UrbanSound8Kvalidation accuracy: 0.143 -> 0.819Improved
Set Gaussian weight-prior means and variational initialization from pretrained deterministic DNN weightsMFVI with MOPED · CIFAR-10validation accuracy: 0.896 -> 0.927 (ResNet-56); 0.878 -> 0.916 (ResNet-20)Improved
Use MOPED weight priors instead of random-initialized priorsMFVI · UCF-101AUPR: 0.0174 -> 0.9186Improved
Use MOPED weight priors instead of random-initialized priorsMFVI · UrbanSound8KAUROC: 0.551 -> 0.9811Improved
Use MOPED weight priors instead of random-initialized priorsMFVI · MNISTAUROC: 0.9999 -> 0.9999No clear effect
Use MOPED weight priors instead of random-initialized priorsMFVI · Diabetic Retinopathyaccuracy at 50% data retained: 0.881 -> 0.937Improved
Use MOPED weight priors instead of random-initialized priorsMFVI · Diabetic RetinopathyAUC at 100% data retained: 0.821 -> 0.883Improved
Use MOPED weight priors instead of random-initialized priorsMFVI · Diabetic Retinopathyaccuracy at 100% data retained: 0.843 -> 0.857Improved
Initialize approximate-posterior variance using a weight-scaled perturbation factor δ (ablation)MOPED · CIFAR-10 and UCF-101AUPRMixed

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