Graph Representation Learning Network via Adaptive Sampling
This work addresses scalability and efficiency issues in graph neural networks for researchers and practitioners working with large-scale graph data, representing an incremental improvement over existing methods.
The paper tackles the scalability limitations of Graph Attention Networks (GAT) on large dense graphs and the neighbor feature combination challenge in GraphSAGE, proposing a new architecture that uses adaptive sampling based on weighted multi-step transition probabilities. Experiments on multiple graph benchmarks like Cora, Citeseer, Pubmed, PPI, Twitter, and YouTube show comparable or better results in transductive and inductive settings.
Graph Attention Network (GAT) and GraphSAGE are neural network architectures that operate on graph-structured data and have been widely studied for link prediction and node classification. One challenge raised by GraphSAGE is how to smartly combine neighbour features based on graph structure. GAT handles this problem through attention, however the challenge with GAT is its scalability over large and dense graphs. In this work, we proposed a new architecture to address these issues that is more efficient and is capable of incorporating different edge type information. It generates node representations by attending to neighbours sampled from weighted multi-step transition probabilities. We conduct experiments on both transductive and inductive settings. Experiments achieved comparable or better results on several graph benchmarks, including the Cora, Citeseer, Pubmed, PPI, Twitter, and YouTube datasets.