AI Paper Atlas

Research map / Learning theory and algorithms / Clustering

Heterogeneous Tensor Decomposition for Clustering via Manifold Optimization

Yanfeng Sun, Junbin Gao, Xia Hong, Bamdev Mishra, Baocai Yin

TPAMI 2016 · Learning theory and algorithms › Clustering · approach: Subspace clustering · 63 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 HOSVD initialization I instead of random initialization (ablation)PCA-Multinomial · PIE Pose27 (24 classes)accuracy: 0.7120 -> 0.7620Improved
use HOSVD initialization II instead of random initialization (ablation)PCA-Multinomial · PIE Pose27 (24 classes)accuracy: 0.7120 -> 0.7240No clear effect
use HOSVD initialization I instead of II (ablation)PCA-Multinomial · PIE Pose27 (24 classes)NMI: 0.8132 -> 0.8199No clear effect
vary tensor core size from 4 to 10, choosing 8 (ablation)PCA-Multinomial · CBCL facesaccuracyMixed
use PCA-Multinomial instead of DRCC with 2 clustersPCA-Multinomial · CBCL facesaccuracy: 0.7352 -> 0.6208Hurt
use PCA-Multinomial instead of PCA-BPCA-Multinomial · CBCL facesaccuracy: 0.6394 -> 0.6208Hurt
use PCA-Multinomial instead of MM at K=10PCA-Multinomial · MNISTaccuracy: 0.4980 -> 0.5126Improved
use PCA-Multinomial instead of MM at K=3PCA-Multinomial · MNISTaccuracy: 0.9173 -> 0.9114No clear effect
use PCA-Multinomial instead of MM at K=68PCA-Multinomial · PIEaccuracy: 0.6088 -> 0.6533Improved
use PCA-Multinomial instead of MM at K=8PCA-Multinomial · PIEaccuracy: 0.6690 -> 0.6523Hurt
use PCA-Multinomial instead of ALSPCA-Multinomial · ORLaccuracy: 0.6820 -> 0.7340Improved
use PCA-Multinomial instead of ALSPCA-Multinomial · YaleB extendedaccuracy: 0.2270 -> 0.3293Improved

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