Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Consistent Semi-Supervised Graph Regularization for High Dimensional Data

About

Semi-supervised Laplacian regularization, a standard graph-based approach for learning from both labelled and unlabelled data, was recently demonstrated to have an insignificant high dimensional learning efficiency with respect to unlabelled data (Mai and Couillet 2018), causing it to be outperformed by its unsupervised counterpart, spectral clustering, given sufficient unlabelled data. Following a detailed discussion on the origin of this inconsistency problem, a novel regularization approach involving centering operation is proposed as solution, supported by both theoretical analysis and empirical results.

Xiaoyi Mai, Romain Couillet• 2020

Related benchmarks

TaskDatasetResultRank
Image ClassificationF-MNIST
Accuracy60.3
139
ClassificationCOIL-20
Accuracy0.769
96
Cancer ClassificationTCGA-BRCA
Accuracy62.5
83
ClassificationTCGA-LUCA
Accuracy88.4
18
Showing 4 of 4 rows

Other info

Follow for update