AI Paper Atlas

Research map / Probabilistic and scientific modeling / Bayesian optimization

GemBag: Group Estimation of Multiple Bayesian Graphical Models

Xinming Yang, Lingrui Gan, Naveen N. Narisetty, Feng Liang

JMLR 2021 · Probabilistic and scientific modeling › Bayesian optimization · approach: Local and high-dimensional Bayesian optimization

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 GemBag (alpha=1) instead of BAGUSjoint graphical model estimation · scale-free network, n=100, p=100, rho=0.25MCC: 0.801 -> 0.939Improved
use GemBag (alpha=1) instead of BAGUSjoint graphical model estimation · circle model, n=200, p=50MCC: 0.892 -> 0.999Improved
use GemBag (alpha=1) instead of Pooled with small samplesjoint graphical model estimation · AR(2) model, n=50, p=50MCC: 0.771 -> 0.721Hurt
use GemBag instead of Pooled with larger samplesjoint graphical model estimation · AR(2) model, n=200, p=50MCC: 0.751 -> 0.986Improved
set sharing parameter alpha=n instead of alpha=1 (ablation)GemBag · scale-free network, n=100, p=100, rho=0.25MCC: 0.939 -> 0.880Hurt
set alpha=sqrt(n) instead of alpha=1 (ablation)GemBag · AR(2) model, n=50, p=50MCC: 0.721 -> 0.783Improved
set alpha=sqrt(n) instead of alpha=1 (ablation)GemBag · AR(2) model, n=200, p=50MCC: 0.986 -> 0.981No clear effect
increase discrepancy between class sparsity patterns rho from 0.25 to 1 (ablation)GemBag (alpha=1) · scale-free network, n=100, p=100MCC: 0.939 -> 0.686Hurt
increase dimension p from 100 to 200 (ablation)GemBag (alpha=1) · scale-free network, n=100, rho=0.25MCC: 0.939 -> 0.909Hurt
use GemBag instead of GGL for computationjoint graphical model estimation · nearest-neighbor network, n=100, p=200computation time (s): 60.421 -> 27.867Improved
use alpha=n instead of alpha=1 (ablation)GemBag · nearest-neighbor network, n=100, p=200computation time (s): 27.867 -> 42.479Hurt
use GemBag instead of BAGUS on bikeshare networksnetwork estimation · Capital Bikeshare registered 2016number of edges: 1042 -> 562Improved

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