A Self-Attention Network based Node Embedding Model
This addresses a critical limitation in graph learning for real-world scenarios where new nodes frequently appear, though it appears incremental as it builds on existing self-attention and random walk methods.
The paper tackles the problem of generating embeddings for unseen nodes in graph networks, which is common in practical applications and affects downstream tasks like node classification. They propose SANNE, an unsupervised model using transformer self-attention on random walks, and it achieves state-of-the-art results on benchmark datasets for node classification.
Despite several signs of progress have been made recently, limited research has been conducted for an inductive setting where embeddings are required for newly unseen nodes -- a setting encountered commonly in practical applications of deep learning for graph networks. This significantly affects the performances of downstream tasks such as node classification, link prediction or community extraction. To this end, we propose SANNE -- a novel unsupervised embedding model -- whose central idea is to employ a transformer self-attention network to iteratively aggregate vector representations of nodes in random walks. Our SANNE aims to produce plausible embeddings not only for present nodes, but also for newly unseen nodes. Experimental results show that the proposed SANNE obtains state-of-the-art results for the node classification task on well-known benchmark datasets.