Abstract:Graph-based solutions have achieved state-of-the-art performance on unsupervised multi-view feature selection. However, existing methods generally characterize the sample similarities first by constructing multiple fixed graphs with manually determined parameters, and then perform the feature selection on a composite one. They will suffer from two severe problems: (1) The fixed graphs may be unreliable as the raw multi-view features usually contain adverse noises and cannot accurately capture the intrinsic sample relations. (2) The graph construction and feature selection are separate and independent, the two-step learning may lead to sub-optimal performance. To tackle these problems, in this paper, we propose an effective unsupervised multi-view feature selection method, dubbed as Nonnegative Structured Graph Learning (NSGL). Specifically, we develop a unified learning framework, which directly learns the structured graph from the raw features by imposing a rank constraint, and simultaneously performs adaptive feature selection with exploiting the complementarity of multi-view features. Besides, we introduce the pseudo label learning to extract the discriminative semantic information in unsupervised scenarios and steer the graph learning process. The informative features are finally selected by forcing the feature selection matrix to be sparse in rows with sparse regression. To solve the challenging optimization problem, we first transform the formulated problem into an equivalent one that can be tackled more easily, and then develop an efficient alternate optimization algorithm guaranteed with convergence to calculate the solution iteratively. Extensive experiments on several widely tested benchmarks demonstrate the superiority of NSGL compared with several state-of-the-art approaches. (C) 2020 Elsevier B.V. All rights reserved.

Multi-Scale Structure-Guided Graph Generation for Multi-View Semi-Supervised Classification

Generative Essential Graph Convolutional Network for Multi-View Semi-Supervised Classification

Semi-Supervised and Self-Supervised Classification with Multi-View Graph Neural Networks

Multi-Scale Graph Convolutional Network with Spectral Graph Wavelet Frame.

Dual Graph Convolutional Networks for Graph-Based Semi-Supervised Classification

Multi-View Clustering and Semi-Supervised Classification with Adaptive Neighbours

Structure fusion based on graph convolutional networks for semi-supervised classification

Multi-view Semi-Supervised Web Image Classification Via Co-Graph

Joint learning of feature and topology for multi-view graph convolutional network

Semi-supervised multi-view graph convolutional networks with application to webpage classification

Heterogeneous graph convolutional network for multi-view semi-supervised classification

Interpretable Graph Convolutional Network for Multi-View Semi-Supervised Learning

Deep Graph-Convolutional Generative Adversarial Network for Semi-Supervised Learning on Graphs

Semisupervised Graph Neural Networks for Graph Classification

Multi-graph Fusion Graph Convolutional Networks with pseudo-label supervision

Multi-scale Graph Clustering Network

Multi-view feature selection via Nonnegative Structured Graph Learning

Attention-based stackable graph convolutional network for multi-view learning

Multi-scale Graph Convolutional Networks with Self-Attention

Learnable Graph Convolutional Network and Feature Fusion for Multi-view Learning

Multi-view Graph Convolutional Networks with Differentiable Node Selection