AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Learning Local Metrics and Influential Regions for Classification

Mingzhi Dong, Yujiang Wang, Xiaochen Yang, Jing‐Hao Xue

TPAMI 2019 · Learning theory and algorithms › Prediction and decision losses · approach: Metric and ordinal embedding · 23 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
use influential-region local metrics instead of R2LMLR2LML · Fourclassaccuracy (%): 76.12 -> 79.71Improved
use influential-region local metrics instead of R2LMLR2LML · Planningaccuracy (%): 63.84 -> 67.40Improved
use influential-region local metrics instead of R2LMLR2LML · Monk1accuracy (%): 89.24 -> 96.46Improved
use influential-region local metrics instead of R2LMLR2LML · Liverdisordersaccuracy (%): 66.81 -> 65.29Hurt
use influential-region local metrics instead of R2LMLR2LML · Heartaccuracy (%): 82.04 -> 81.67No clear effect
use influential-region local metrics instead of GMMLGMML · Breastcanceraccuracy (%): 97.26 -> 97.15No clear effect
use local metrics instead of single global metric (LMNN)LMNN · Diabetesaccuracy (%): 69.16 -> 75.19Improved
use local metrics instead of SCMLSCML · Monk1accuracy (%): 97.53 -> 96.46No clear effect
use local metrics across all 14 datasetsbest competitor · 14 UCI/LibSVM binary datasetsnumber of best results: 8 of 14 vs at most 4 for othersImproved

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