AI Paper Atlas

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

Gradients Weights improve Regression and Classification

Samory Kpotufe, Abdeslam Boularias, Thomas Schultz, Kyoungok Kim

JMLR 2016 · Learning theory and algorithms › Kernels and regression theory · approach: High-dimensional asymptotic generalization · 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
add gradient weights (squared) to k-NN regressionk-NN · Barrett 1normalized MSE: 0.41 ± 0.02 -> 0.21 ± 0.02Improved
add gradient weights (squared) to kernel regressionKR · Housingnormalized MSE: 0.37 ± 0.08 -> 0.21 ± 0.04Improved
add gradient weights (squared) to k-NN regressionk-NN · Telecomnormalized MSE: 0.13 ± 0.02 -> 0.34 ± 0.05Hurt
add gradient weights (squared) to kernel regressionKR · Telecomnormalized MSE: 0.30 ± 0.02 -> 0.37 ± 0.08Hurt
normalize features by training standard deviationKR · Barrett 1normalized MSE: 0.98 ± 0.03 -> 0.50 ± 0.02Improved
use squared gradient weights instead of non-squared (ablation)KR · Barrett 1normalized MSE: 0.38 ± 0.03 -> 0.30 ± 0.03Improved
add gradient weights (squared) to k-NN classifierk-NN · IJCNNerror rate: 0.0629 ± 0.0038 -> 0.0482 ± 0.0060Improved
add gradient weights (squared) to k-NN classifierk-NN · MAGIC Gammaerror rate: 0.1884 ± 0.0118 -> 0.1875 ± 0.0092No clear effect
add gradient weights (squared) to k-NN classifierk-NN · Shuttleerror rate: 0.0031 ± 0.0011 -> 0.0019 ± 0.0008Improved
add gradient weights (squared) to epsilon-NN classifiere-NN · Covertypeerror rate: 0.28 ± 0.01 -> 0.25 ± 0.01Improved
replace GW with ReliefF weights for k-NN (ablation)k-NN · Shuttleerror improvement over k-NN: 0.0006±0.0011 -> -0.0019±0.0024Hurt
replace GW with ReliefF weights for k-NN (ablation)k-NN · Covertypeerror improvement over k-NN: 0.0113±0.0067 -> 0.0229±0.0075Improved

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