AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

The Maximum Separation Subspace in Sufficient Dimension Reduction with Categorical Response

Xin Zhang, Qing Mai, Hui Zou

JMLR 2020 · Learning theory and algorithms › Sample complexity · approach: Density and property estimation

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 MASES instead of SIRSDR estimators · Inverse models MDA-1angle (degrees): 64.7 -> 16.3Improved
use MASES instead of SIRSDR estimators · Inverse model MDA-2angle (degrees): 53.3 -> 10.4Improved
use MASES instead of SIR on correlated-predictor LDA modelsSDR estimators · LDA-1 Cor.angle (degrees): 52.7 -> 27.0Improved
use MASES instead of PSVM/PRE when classes are almost completely separated and Sigma = ISDR estimators · LDA-2 Ind.angle (degrees): 23.3 -> 34.4Hurt
use MASES instead of SIR on forward single-index modelsSDR estimators · SIM-1angle (degrees): 80.6 -> 37.8Improved
use MASES instead of logistic regression with independent predictorsSDR estimators · LR-Ind.angle (degrees): 20.1 -> 22.2Hurt
use MASES instead of SIR with correlated predictors in logistic modelSDR estimators · LR-Cor.angle (degrees): 50.1 -> 29.2Improved
use MASES instead of the best competitor on two-dimensional modelsSDR estimators · MIM-3Frobenius estimation error: 0.79 -> 0.29Improved
use MASES instead of SIR on two-dimensional LDA with correlated predictorsSDR estimators · LDA Cor. (2D)Frobenius estimation error: 1.62 -> 0.41Improved
apply MASES dimension selection procedureMASES · Two-dimensional models, 100 replicatescorrect dimension selection: 65 of 100 on MIM-1 vs 100 on LDA Ind.Mixed
use MASES instead of SAVE/DR subspaces for class separationSDR estimators · Birds, planes and cars dataoverall Hellinger distance: 0.133 -> 0.982Improved

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