AI Paper Atlas

Research map / Learning theory and algorithms / Kernels and regression theory

On Sufficient Graphical Models

Bing Li, Kyongwon Kim

JMLR 2024 · Learning theory and algorithms › Kernels and regression theory · approach: Kernel regression under shift · 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
add nonlinear sufficient dimension reduction step (vs Naive conjoined conditional covariance operator) (ablation)Sufficient graphical model · DREAM 4 Network 1AUC: 0.78 -> 0.85Improved
add nonlinear sufficient dimension reduction step (vs Naive) (ablation)Sufficient graphical model · DREAM 4 Network 5AUC: 0.71 -> 0.79Improved
add nonlinear sufficient dimension reduction step (vs Naive) (ablation)Sufficient graphical model · DREAM 4 Network 2AUC: 0.76 -> 0.81Improved
add dimension reduction (vs fully nonparametric methods) in high-dimensional setting p=200 > nSufficient graphical model · Simulation Models III and IV (p=200, n=50, 100)averaged ROC curveImproved
use sufficient graphical model instead of Gaussian or copula Gaussian methods on non-Gaussian dataSufficient graphical model · Simulation Models I and IIaveraged ROC curveImproved
use sufficient graphical model instead of additive conditional independence methodSufficient graphical model · Simulation Model I (additive) vs Model II (non-additive)averaged ROC curveMixed
use sufficient graphical model instead of Gaussian graphical model methods on Gaussian dataSufficient graphical model · Simulation Model V (Gaussian, p=20)averaged ROC curveHurt
use d_ij=2 instead of d_ij=1 for dimension of reduced space (ablation)Sufficient graphical model · DREAM 4AUCNo clear effect
use d_ij=1 instead of d_ij=2 for dimension of reduced space (ablation)Sufficient graphical model · Simulation Models I-IIaveraged ROC curveNo clear effect
sufficient graphical model vs DREAM 4 championSufficient graphical model · DREAM 4 Network 1AUC: 0.91 -> 0.85Hurt
sufficient graphical model vs DREAM 4 championSufficient graphical model · DREAM 4 Network 5AUC: 0.75 -> 0.79Improved

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