AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Similarity Learning via Kernel Preserving Embedding

Zhao Kang, Yiwei Lu, Yuanzhang Su, Changsheng Li, Zenglin Xu

AAAI 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 42 citations (OpenAlex)

Read the paper

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
reconstruct kernel matrix instead of raw data (SLKE-R vs LRR)low-rank self-expression · 9 clustering datasetsaccuracy: 8.92% average accuracy gainImproved
reconstruct kernel matrix instead of raw data (SLKE-S vs SSC)sparse self-expression · 9 clustering datasetsaccuracy: 8.76% average accuracy gainImproved
kernel-preserving similarity vs raw-data LRRLRR · YALEaccuracy: 61.21 -> 66.24Improved
kernel-preserving similarity vs raw-data SSCSSC · TDT2accuracy: 23.45 -> 74.92Improved
SLKE vs adaptive-neighbor CANSLKE-R · COIL20accuracy: 84.58 -> 84.03Hurt
SLKE vs adaptive-neighbor CANSLKE-R · TR41accuracy: 62.87 -> 74.37Improved
SLKE vs TLSCSLKE-R · TDT2accuracy: 55.74 -> 74.92Improved
low-rank (SLKE-R) vs sparse (SLKE-S) regularizerSLKE · ORL / BAaccuracy: ORL 77.00 vs 74.75; BA 50.74 vs 44.37Mixed
best kernel vs average over 12 kernelsSLKE-R · YALEaccuracy: 51.28 -> 66.24Improved
vary gamma from 1e-6 to 1e-3 (ablation)SLKE · YALE and JAFFE, Gaussian kernel t=100accuracy / NMINo clear effect
Wilcoxon test of SLKE-R vs SCSLKE-R · all 9 datasetsaccuracy: p = .0039Improved
Wilcoxon test of SLKE-R vs LRRSLKE-R · all 9 datasetsaccuracy: p = .0977No clear effect

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