graph2vec: Learning Distributed Representations of Graphs

TLDR

Graph representation learning has largely focused on substructures, yet many tasks require fixed‑length vectors for entire graphs, and existing graph kernels rely on handcrafted features that limit generalization. This work introduces graph2vec, a neural embedding framework that learns data‑driven, fixed‑length representations for graphs of arbitrary size. Graph2vec learns embeddings in an unsupervised, task‑agnostic manner, enabling their use for downstream classification, clustering, and as seeds for supervised learning. Experiments on benchmark and large real‑world datasets demonstrate that graph2vec significantly outperforms substructure‑based methods in classification and clustering while remaining competitive with state‑of‑the‑art graph kernels.

Abstract

Recent works on representation learning for graph structured data predominantly focus on learning distributed representations of graph substructures such as nodes and subgraphs. However, many graph analytics tasks such as graph classification and clustering require representing entire graphs as fixed length feature vectors. While the aforementioned approaches are naturally unequipped to learn such representations, graph kernels remain as the most effective way of obtaining them. However, these graph kernels use handcrafted features (e.g., shortest paths, graphlets, etc.) and hence are hampered by problems such as poor generalization. To address this limitation, in this work, we propose a neural embedding framework named graph2vec to learn data-driven distributed representations of arbitrary sized graphs. graph2vec's embeddings are learnt in an unsupervised manner and are task agnostic. Hence, they could be used for any downstream task such as graph classification, clustering and even seeding supervised representation learning approaches. Our experiments on several benchmark and large real-world datasets show that graph2vec achieves significant improvements in classification and clustering accuracies over substructure representation learning approaches and are competitive with state-of-the-art graph kernels.