AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Deep Single-Index Fréchet Regression

Muqing Cui, Yidong Zhou, Su I Iao, Hans-Georg Müller

ICML 2026 · Learning theory and algorithms › Matrix and tensor methods · approach: Geometry of low-rank and SPD structure · 0 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
replace linear single-index (IFR) with DNN-learned index (DeSI)Frechet regression on networks · Simulation Networks (n=1000)mean prediction error: 0.4927 -> 0.0531Improved
replace linear single-index (IFR) with DNN-learned index (DeSI)Frechet regression on SPD matrices · Simulation SPD (n=1000)mean prediction error: 0.0101 -> 0.0057Improved
use DeSI instead of global Frechet regression at small sample sizeFrechet regression on distributions (linear) · Simulation Distributions linear (n=200)mean prediction error: GFR 0.0744 vs DeSI 0.1582Hurt
increase sample size from 200 to 1000DeSI · Simulation Networksmean prediction error: 0.2214 -> 0.0531Improved
learn index with DeSI instead of IFRindex direction estimation · Simulation Networks (n=1000)index estimation error: 0.1744 -> 0.0023Improved
learn index with DeSI instead of IFRindex direction estimation · Simulation Distributions quadratic (n=1000)index estimation error: 0.2211 -> 0.0220Improved
DeSI instead of IFR and GFR on real mood datacompositional mood regression · New Jersey compositional mood dataMPE: IFR 0.4947 -> 0.4658; GFR 0.6229 -> 0.4658Improved

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