AI Paper Atlas

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

Scalable Gaussian-process regression and variable selection using Vecchia approximations

Jian Cao, Joseph Guinness, Marc G. Genton, Matthias Katzfuss

JMLR 2022 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: PAC-Bayes and decoupled GP learning · 8 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 VGPR instead of Lassosparse GP regression · Slice-Itest RMSE: 0.44 -> 0.32Improved
use VGPR instead of Lassocovariate selection · Slice-Inumber of selected covariates: 792 -> 49Improved
use VGPR instead of Lassocovariate selection · CASP-Ifalse positive ratio: 96% -> 0%Improved
use VGPR instead of Treeregression · Piston-Dtest RMSE: 0.71 -> 0.00Improved
use VGPR instead of Lassoregression · Temptest RMSE: 0.28 -> 0.29No clear effect
use VGPR instead of Lassocovariate selection · Tempnumber of selected covariates: 84 -> 6Improved
add correlated artificial covariatesVGPR · CASPtest RMSE: 0.78 -> 0.75Improved
use proposed Vecchia mini-batch subsampling instead of comparison methodsSR-gradient estimator · simulated GP, n=10^4absolute bias and RMSEImproved
add mini-batch subsampling with iterative adaptive bridge penaltyVGPR regularization path · simulated GPcomputation time: reduced by more than 90%Improved
use sequential warm-start addition of covariates (FWD, VGPR) instead of full-d optimization (PGPR, Fisher)GP regression · simulated d=1,000RMSEImproved
use VGPR instead of other GP methodsvariable selection · simulation n>=5,000false positive ratio: zero FPos when n >= 5,000Improved
use FWD instead of VGPRGP regression · simulation n=500RMSENo clear effect

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