AI Paper Atlas

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

Quantile Graphical Models: a Bayesian Approach

Nilabja Guha, Veera Baladandayuthapani, Bani K. Mallick

JMLR 2020 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Nonparametric and PAC-Bayes Bayesian learning

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 quantile variational Bayes (tau=.5) instead of GGM (MB)graph estimation · Example 1(a) (P=30, n=400, nonlinear/heavy-tailed)FDR: 5.21 -> 0.28Improved
use quantile variational Bayes (tau=.5) instead of GGM (GLASSO)graph estimation · Example 1(a) (P=30, n=400, nonlinear/heavy-tailed)FDR: 14.78 -> 0.28Improved
add quantiles .3 and .7 to tau={.5} (ablation)QVB · Example 1(a) (P=30, n=400, nonlinear/heavy-tailed)e1 / FDR: e1 0.32 -> 0.17; FDR 0.28 -> 0.58Mixed
use higher quantile tau=.9 instead of tau=.3 (ablation)QVB-MCMC · Example 3a, n=200e_n: 3.88 -> 1.65Improved
increase sample size from n=200 to n=500 (ablation)QVB-MCMC tau=.9 · Example 3ae_n: 1.65 -> 0.29Improved
use quantiles .2,.5,.8 instead of tau=.5 alone (ablation)QVB · Example 3a.ii (hierarchical scale)detecting connections to scale variable X16Improved
densify quantile grid from step 0.1 to 0.025 (ablation)QVB · Example 3b (granularity of quantile grid)FDR: 0.68 -> 0.77No clear effect
add quantiles .3 and .7 to tau={.5} (ablation)QVB · Example 3b (granularity of quantile grid)e1: 0.30 -> 0.11Improved
use variational Bayes instead of MCMCQVB · Example 1 setup, n=400runtime speedup: 170 and 134 times fasterImproved
vary prior variance v from 1 to 100 (ablation)QVB · Example 1(a)graph estimateNo 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