InstantEmbedding: Efficient Local Node Representations

Abstract

In this paper, we introduce InstantEmbedding, an efficient method for generating single-node representations using local PageRank computations. We theoretically prove that our approach produces globally consistent representations in sublinear time. We demonstrate this empirically by conducting extensive experiments on real-world datasets with over a billion edges. Our experiments confirm that InstantEmbedding requires drastically less computation time (over 9,000 times faster) and less memory (by over 8,000 times) to produce a single node's embedding than traditional methods including DeepWalk, node2vec, VERSE, and FastRP. We also show that our method produces high quality representations, demonstrating results that meet or exceed the state of the art for unsupervised representation learning on tasks like node classification and link prediction.

arXiv comment23 pages, 9 figures

References in corpus (7)

• Scikit-learn: Machine Learning in Python
• Semi-Supervised Classification with Graph Convolutional Networks
• VERSE: Versatile Graph Embeddings from Similarity Measures
• A Tutorial on Network Embeddings
• FAST-PPR: Scaling Personalized PageRank Estimation for Large Graphs
• On Node Features for Graph Neural Networks
• Billion-scale Network Embedding with Iterative Random Projection

Cited by in corpus (3)

• FREDE: Anytime Graph Embeddings
• QUINT: Node embedding using network hashing
• Subset Node Representation Learning over Large Dynamic Graphs