AI Paper Atlas

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

Generalized Variational Inference in Function Spaces: Gaussian Measures meet Bayesian Deep Learning

Veit David Wild, Robert Hu, Dino Sejdinovic

NeurIPS 2022 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Sparse variational Gaussian processes · 1 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 a DNN variational mean with the SVGP covariance instead of the SVGP mean and covarianceGWI · UCI regression datasetsmean rank: 5.5 -> 2.06Improved
Use GWI-net rather than the strongest listed baselineimage classification models · Fashion MNISTaccuracy: 91.60$\pm$0.14 -> 93.25 $\pm$0.09Improved
Use GWI-net rather than the strongest listed baselineimage classification models · Fashion MNISTNLL: 0.254$\pm$0.05 -> 0.250 $\pm$0.00Improved
Use GWI-net rather than the strongest listed baselineimage classification models · Fashion MNISTOOD-AUC: 0.956 $\pm$0.06 -> 0.959 $\pm$0.01No clear effect
Use GWI-net rather than the strongest listed baselineimage classification models · CIFAR-10accuracy: 77.69 $\pm$0.64 -> 83.82 $\pm$0.00Improved
Use GWI-net rather than the strongest listed baselineimage classification models · CIFAR-10NLL: 0.661$\pm$0.00 -> 0.553 $\pm$0.00Improved
Use GWI-net rather than the strongest listed baselineimage classification models · CIFAR-10OOD-AUC: 0.883 $\pm$0.04 -> 0.618 $\pm$0.00Hurt

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