AI Paper Atlas

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

Learning Data-adaptive Non-parametric Kernels

Fanghui Liu, Xiaolin Huang, Chen Gong, Jie Yang, Li Li

JMLR 2020 · Learning theory and algorithms › Kernels and regression theory · approach: RKHS kernel methods

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
learn data-adaptive low-rank bounded matrix F on Gaussian kernel (DANK) vs SVM-CVSVM with Gaussian kernel · UCI diabetictest accuracy (%): 73.0 -> 81.2Improved
learn DANK matrix FSVM with Gaussian kernel · UCI hearttest accuracy (%): 81.9 -> 87.9Improved
learn DANK matrix FSVM with Gaussian kernel · UCI glasstest accuracy (%): 69.8 -> 74.5Improved
learn DANK matrix FSVM with Gaussian kernel · UCI monks3test accuracy (%): 93.0 -> 93.0No clear effect
learn DANK matrix FSVM with Gaussian kernel · UCI monks2, sonar, winetest accuracy (%): 85.8 -> 86.7Mixed
add bounded constraint and low-rank structure on F (DANK vs unconstrained MKL-uniform) (ablation)MKL-uniform · UCI diabetictest accuracy (%): 61.9 -> 81.2Improved
add bounded and low-rank constraints on F (DANK vs MKL-uniform) (ablation)MKL-uniform · UCI glasstest accuracy (%): 53.0 -> 74.5Improved
add bounded constraint and low-rank structure to KNPL (DANK) (ablation)KNPL · UCI monks3test accuracy (%): 88.7 -> 93.0Improved
add bounded constraint and low-rank structure to KNPL (DANK) (ablation)KNPL · UCI monks2test accuracy (%): 83.3 -> 86.7Improved
add bounded constraint and low-rank structure to KNPL (DANK) (ablation)KNPL · UCI specttest accuracy (%): 79.7 -> 78.9Hurt
use DANK instead of LogDetLogDet · UCI monks1test accuracy (%): 86.6 -> 83.6Hurt
use DANK instead of LogDetLogDet · UCI hearttest accuracy (%): 80.6 -> 87.9Improved

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