Almost exact recovery in noisy semi-supervised learning

arXiv2007.14717MaRDI QIDQ6346087

Author name not available (Why is that?)

Publication date: 29 July 2020

Abstract: Graph-based semi-supervised learning methods combine the graph structure and labeled data to classify unlabeled data. In this work, we study the effect of a noisy oracle on classification. In particular, we derive the Maximum A Posteriori (MAP) estimator for clustering a Degree Corrected Stochastic Block Model (DC-SBM) when a noisy oracle reveals a fraction of the labels. We then propose an algorithm derived from a continuous relaxation of the MAP, and we establish its consistency. Numerical experiments show that our approach achieves promising performance on synthetic and real data sets, even in the case of very noisy labeled data.

Has companion code repository: https://github.com/mdreveton/ssl-sbm

Mathematics Subject Classification ID

No records found.

This page was built for publication: Almost exact recovery in noisy semi-supervised learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6346087)