Semi-supervised learning with regularized Laplacian

DOI10.1080/10556788.2016.1193176zbMATH Open1365.68366arXiv1508.04906OpenAlexW2218183746MaRDI QIDQ5268919FDOQ5268919

A. Mishenin, Konstantin Avrachenkov, Pavel Chebotarev

Publication date: 21 June 2017

Published in: Optimization Methods \& Software (Search for Journal in Brave)

Abstract: We study a semi-supervised learning method based on the similarity graph and RegularizedLaplacian. We give convenient optimization formulation of the Regularized Laplacian method and establishits various properties. In particular, we show that the kernel of the methodcan be interpreted in terms of discrete and continuous time random walks and possesses several importantproperties of proximity measures. Both optimization and linear algebra methods can be used for efficientcomputation of the classification functions. We demonstrate on numerical examples that theRegularized Laplacian method is competitive with respect to the other state of the art semi-supervisedlearning methods.

Full work available at URL: https://arxiv.org/abs/1508.04906

Recommendations

• Properly-weighted graph Laplacian for semi-supervised learning
• The performance of semi-supervised Laplacian regularized regression with the least square loss
• Learning Theory
• Generalization errors of Laplacian regularized least squares regression
• Semi-supervised learning on Riemannian manifolds

zbMATH Keywords

semi-supervised learningproximity measuregraph-based learningregularized LaplacianWikipedia article classification

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Graphs and linear algebra (matrices, eigenvalues, etc.) (05C50) Random walks on graphs (05C81)

Cites Work

• On certain linear mappings between inner-product and squared-distance matrices
• Distances in Weighted Trees and Group Inverse of Laplacian Matrices
• Discrete Dynamic Programming
• Geometry of cuts and metrics
• Title not available (Why is that?)
• On Some Ridge Regression Estimators: An Empirical Comparisons
• The graph bottleneck identity
• Spanning forests of a digraph and their applications
• An experimental investigation of kernels on graphs for collaborative recommendation and semisupervised classification
• Discrete Green's functions
• Spanning forests and the golden ratio
• Aggregation of preferences by the generalized row sum method

Cited In (21)

• Analysis of Relaxation Time in Random Walk with Jumps
• Learning Theory
• Semi-Supervised Bilinear Subspace Learning
• Margin-based semi-supervised learning using Apollonius circle
• An efficient kernel learning algorithm for semisupervised regression problems
• ELECTRIC NETWORK CLASSIFIERS FOR SEMI-SUPERVISED LEARNING ON GRAPHS
• Title not available (Why is that?)
• Learning from normalized local and global discriminative information for semi-supervised regression and dimensionality reduction
• Semi-supervised learning on Riemannian manifolds
• Regularized semi-supervised least squares regression with dependent samples
• Loop-erased partitioning of a graph: mean-field analysis
• Loop-erased partitioning via parametric spanning trees: monotonicities \& 1D-scaling
• Title not available (Why is that?)
• Semi-supervised learning with the help of Parzen windows
• Rates of convergence for Laplacian semi-supervised learning with low labeling rates
• Semi-supervised learning based on high density region estimation
• Laplacian-optimized diffusion for semi-supervised learning
• Graph-Based Semi-Supervised Learning and Spectral Kernel Design
• Guaranteed Classification via Regularized Similarity Learning
• Semantic-based regularization for learning and inference
• Similarities on graphs: kernels versus proximity measures

Uses Software

• GraphBase
• CUDA

This page was built for publication: Semi-supervised learning with regularized Laplacian

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5268919)