AI Paper Atlas

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

Bayesian Spiked Laplacian Graphs

Leo L Duan, George Michailidis, Mingzhou Ding

JMLR 2023 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Priors and functional variational uncertainty

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 spiked Laplacian instead of direct decomposition of observed Laplacianspectral community detection · single graph, spectral gap 0.05NMI: 0.42 -> 0.58Improved
use spiked Laplacian instead of direct decomposition of observed Laplacianspectral community detection · single graph, spectral gap 0.1NMI: 0.78 -> 0.88Improved
use spiked Laplacian instead of direct decomposition of observed Laplacianspectral community detection · single graph, spectral gap 0.01NMI: 0.36 -> 0.40Improved
use spiked Laplacian model instead of stochastic block modelcommunity detection · n=100NMI: 0.65 -> 0.84Improved
use spiked Laplacian model instead of Bayesian SBMcommunity detection · n=100NMI: 0.70 -> 0.84Improved
use spiked Laplacian model instead of SBMcommunity detection · n=500NMI: 0.87 -> 0.86No clear effect
use spiked Laplacian model instead of SBMcommunity detection · n=300NMI: 0.84 -> 0.90Improved
use spiked Laplacian graphs instead of averaging graphs then SBMheterogeneous graph collection · 500 simulated graphsNMI: 0.21 -> 0.85Improved
use spiked Laplacian graphs instead of clustering graphs then SBMsheterogeneous graph collection · 500 simulated graphsNMI: 0.67 -> 0.85Improved
use spiked Laplacian graphs instead of individual SBMsheterogeneous graph collection · 500 simulated graphsNMI: 0.45 -> 0.85Improved
use individual SBMs instead of spiked Laplacian graphsheterogeneous graph collection · 500 simulated graphsRMSE (x10^-3): 1.9 -> 1.2Improved
over-specify T (T=10 or T=30 versus true number of eigenvalue components) (ablation)spiked Laplacian model · single graph, three communitiesfirst eigenvalue estimatesNo 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