Anonymous Walk Embeddings

Abstract

The task of representing entire graphs has seen a surge of prominent results, mainly due to learning convolutional neural networks (CNNs) on graph-structured data. While CNNs demonstrate state-of-the-art performance in graph classification task, such methods are supervised and therefore steer away from the original problem of network representation in task-agnostic manner. Here, we coherently propose an approach for embedding entire graphs and show that our feature representations with SVM classifier increase classification accuracy of CNN algorithms and traditional graph kernels. For this we describe a recently discovered graph object, anonymous walk, on which we design task-independent algorithms for learning graph representations in explicit and distributed way. Overall, our work represents a new scalable unsupervised learning of state-of-the-art representations of entire graphs.

ICML 2018

References in corpus (3)

• Efficient Estimation of Word Representations in Vector Space
• Learning Convolutional Neural Networks for Graphs
• graph2vec: Learning Distributed Representations of Graphs

Cited by in corpus (39)

• How Powerful are Graph Neural Networks?
• A Gentle Introduction to Deep Learning for Graphs
• Improving Graph Neural Network Expressivity via Subgraph Isomorphism Counting
• InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization
• Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels
• Second-Order Pooling for Graph Neural Networks
• Learning Backtrackless Aligned-Spatial Graph Convolutional Networks for Graph Classification
• GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training
• Power Normalizations in Fine-grained Image, Few-shot Image and Graph Classification
• Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks
• Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification
• Natural Graph Networks
• SUGAR: Subgraph Neural Network with Reinforcement Pooling and Self-Supervised Mutual Information Mechanism
• Convolutional Kernel Networks for Graph-Structured Data
• Provably Powerful Graph Networks
• Few-Shot Learning on Graphs via Super-Classes based on Graph Spectral Measures
• Learning Universal Graph Neural Network Embeddings With Aid Of Transfer Learning
• Ensemble of Random and Isolation Forests for Graph-Based Intrusion Detection in Containers
• The Impact of Global Structural Information in Graph Neural Networks Applications
• Revisiting 2D Convolutional Neural Networks for Graph-based Applications
• Graph Convolutional Neural Networks via Motif-based Attention
• A Lagrangian Approach to Information Propagation in Graph Neural Networks
• LCS Graph Kernel Based on Wasserstein Distance in Longest Common Subsequence Metric Space
• Quaternion Graph Neural Networks
• Wasserstein Embedding for Graph Learning
• Learning Graph Structure With A Finite-State Automaton Layer
• Capsule Graph Neural Networks with EM Routing
• Learning distributed representations of graphs with Geo2DR
• GraLSP: Graph Neural Networks with Local Structural Patterns
• Robust Hierarchical Graph Classification with Subgraph Attention
• Unsupervised Graph Representation by Periphery and Hierarchical Information Maximization
• G5: A Universal GRAPH-BERT for Graph-to-Graph Transfer and Apocalypse Learning
• Deep Constraint-based Propagation in Graph Neural Networks
• Convergent Graph Solvers
• CGCL: Collaborative Graph Contrastive Learning without Handcrafted Graph Data Augmentations
• Graph embedding using multi-layer adjacent point merging model
• Graph-Preserving Grid Layout: A Simple Graph Drawing Method for Graph Classification using CNNs
• A Literature Review of Recent Graph Embedding Techniques for Biomedical Data
• Graph Classification Based on Skeleton and Component Features