Node-Feature Convolution for Graph Convolutional Networks

Article Computer Science, Artificial Intelligence

GraphAIR: Graph representation learning with neighborhood aggregation and interaction

Fenyu Hu et al.

Summary: Research shows that the coefficients of neighborhood interacting terms in current graph representation learning models are relatively small, making it difficult to capture the complex non-linearity of graph data. To address this, a novel GraphAIR framework is proposed to better model neighborhood interaction in addition to neighborhood aggregation, and it proves to be effective in tasks such as node classification and link prediction.

PATTERN RECOGNITION (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Image super -resolution via channel attention and spatial graph convolutional network

Yue Yang et al.

Summary: In this work, a channel attention and spatial graph convolutional network (CASGCN) is proposed for more powerful feature obtaining and feature correlations modeling by incorporating global spatial and channel inter-dependencies. CASGCN utilizes channel and spatial branches to learn the representation of each image pixel, using attention mechanism and spatial-aware graph to extract informative features and model global self-similar information. The adjacency matrix in spatial-aware graph is dynamically generated via the Gram matrix to model global correlations between pixels and is shared across the whole network without auxiliary parameters. Extensive experiments demonstrate the effectiveness of CASGCN in single image superresolution tasks.

PATTERN RECOGNITION (2021)

Add to Collection

Article Computer Science, Artificial Intelligence