Representation Learning on Graphs with Jumping Knowledge Networks

TL;DR

This paper introduces Jumping Knowledge networks, a flexible architecture for graph representation learning that adaptively combines information from different neighborhood ranges, leading to improved performance across various graph-based tasks.

Contribution

The paper proposes Jumping Knowledge networks, a novel architecture that enhances existing graph neural networks by adaptively aggregating neighborhood information for better structure-aware representations.

Findings

Achieves state-of-the-art results on social, bioinformatics, and citation networks.

Improves performance of GCN, GraphSAGE, and Graph Attention Networks.

Demonstrates the effectiveness of adaptive neighborhood aggregation.

Abstract

Recent deep learning approaches for representation learning on graphs follow a neighborhood aggregation procedure. We analyze some important properties of these models, and propose a strategy to overcome those. In particular, the range of "neighboring" nodes that a node's representation draws from strongly depends on the graph structure, analogous to the spread of a random walk. To adapt to local neighborhood properties and tasks, we explore an architecture -- jumping knowledge (JK) networks -- that flexibly leverages, for each node, different neighborhood ranges to enable better structure-aware representation. In a number of experiments on social, bioinformatics and citation networks, we demonstrate that our model achieves state-of-the-art performance. Furthermore, combining the JK framework with models like Graph Convolutional Networks, GraphSAGE and Graph Attention Networks consistently improves those models' performance.