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.

Graph Convolutional Network Using a Reliability-Based Feature Aggregation Mechanism

Multi-Behavior Enhanced Heterogeneous Graph Convolutional Networks Recommendation Algorithm based on Feature-Interaction

Robust graph learning with graph convolutional network

Deep Graph Neural Networks via Flexible Subgraph Aggregation

GraphAIR: Graph representation learning with neighborhood aggregation and interaction

Graph neural networks via contrast between separation and aggregation for self and neighborhood

Exploiting Neighbor Effect: Conv-Agnostic GNN Framework for Graphs with Heterophily

Multigraph Fusion for Dynamic Graph Convolutional Network

Feature Correlation Aggregation: on the Path to Better Graph Neural Networks

Reliable Data Distillation on Graph Convolutional Network.

Attention-based Graph Convolution Fusing Latent Structures and Multiple Features for Graph Neural Networks

Self-SAGCN: Self-Supervised Semantic Alignment for Graph Convolution Network

Adaptive Graph Convolutional Collaboration Networks for Semi-supervised Classification

AF-GCN: Attribute-Fusing Graph Convolution Network for Recommendation

Facial Expression Recognition on the High Aggregation Subgraphs

Neighbor enhanced graph convolutional networks for node classification and recommendation

Dual Feature Interaction-Based Graph Convolutional Network

MobileGCN applied to low-dimensional node feature learning

Graph Attribute Aggregation Network with Progressive Margin Folding.

Adaptive Sampling Towards Fast Graph Representation Learning

TransGCN:Coupling Transformation Assumptions with Graph Convolutional Networks for Link Prediction