Abstract:Graph convolution networks (GCNs) have achieved remarkable success in processing non-Euclidean data. GCNs update the feature representations of each sample by aggregating the structure information from K -order (layer) neighborhood samples. Existing GCNs variants rely heavily on the K -th layer semantic information with K -order neighborhood information aggregating. However, semantic features from different convolution layers have distinct sample attributes. The single-layer semantic feature is only a one-sided feature representation. Besides, the semantic features of traditional GCNs will be oversmoothing with multi-layer structure information aggregates. In this paper, to solve the above-mentioned problem, we propose adaptive graph convolutional collaboration networks (AGCCNs) for the semi-supervised classification task. AGCCNs can fully use the different scales of discrimination information contained in the different convolutional layers. Specifically, AGCCNs utilize the attention mechanism to learn the relevance (contribution) coefficient of the deep semantic features from different convolution layers for the task, which aims to effectively discriminant their importance. After multiple optimizations, AGCCNs can adaptively learn the robust deep semantic features via the effective semantic fusion process between multi-layer semantic information. Compared with GCNs that only utilize the K -th layer semantic features, AGCCNs make the learned deep semantic features contain richer and more robust semantic information. What is more, our proposed AGCCNs can aggregate the appropriate K -order neighborhood information for each sample, which can relieve the oversmoothing issue of traditional GCNs and better generalize shallow GCNs to more deep layers. Abundant experimental results on several popular datasets demonstrate the superiority of our proposed AGCCNs compared with traditional GCNs.

Example-feature graph convolutional networks for semi-supervised classification

Semi-supervised classification by graph p-Laplacian convolutional networks

Dynamic Graph Learning Convolutional Networks for Semi-supervised Classification

Semi-supervised classification by graph <i>p</i>-Laplacian convolutional networks

Scalable Graph Convolutional Networks with Fast Localized Spectral Filter for Directed Graphs

Fuzzy graph convolutional network for hyperspectral image classification

Adaptive Graph Convolutional Collaboration Networks for Semi-supervised Classification

CNN-Enhanced Graph Convolutional Network With Pixel- and Superpixel-Level Feature Fusion for Hyperspectral Image Classification

HesGCN: Hessian graph convolutional networks for semi-supervised classification

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

Graph Classification via Discriminative Edge Feature Learning

Two-order Graph Convolutional Networks for Semi-Supervised Classification.

High-order Graph Convolutional Networks for Semi-Supervised Classification

Superpixel Spectral–Spatial Feature Fusion Graph Convolution Network for Hyperspectral Image Classification

Automatic Graph Learning Convolutional Networks for Hyperspectral Image Classification

Graph convolutional networks of reconstructed graph structure with constrained Laplacian rank

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

Hyperspectral Image Classification with a Multiscale Fusion-Evolution Graph Convolutional Network Based on a Feature-Spatial Attention Mechanism

HpLapGCN: Hypergraph P-Laplacian Graph Convolutional Networks

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

Semi-Supervised Classification with Graph Convolutional Networks