Revisiting and improving semi-supervised learning: a large dimensional approach

Abstract : The recent work [1] shows that in the big data regime (i.e., numerous high dimensional data), the popular semi-supervised graph regularization, known as semi-supervised Laplacian regularization, fails to effectively extract information from unlabelled data. In response to this problem, we propose in this article an improved approach based on a simple yet fundamental update of the classical method. The effectiveness of the former is supported by both asymptotic results and simulations on finite data sets. Index Terms-semi-supervised learning, large dimensional statistics, random matrix theory.
Document type :
Conference papers
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download
Contributor : Xiaoyi Mai <>
Submitted on : Sunday, May 26, 2019 - 5:10:15 PM
Last modification on : Monday, July 8, 2019 - 12:29:32 PM


Revisiting and improving semi-...
Files produced by the author(s)



Xiaoyi Mai, Romain Couillet. Revisiting and improving semi-supervised learning: a large dimensional approach. ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), May 2019, Brighton, United Kingdom. ⟨10.1109/ICASSP.2019.8683378⟩. ⟨hal-02139979⟩



Record views


Files downloads