Distance-Preserving Graph Embeddings from Random Neural Features

09/09/2019

∙

We present Graph Random Neural Features (GRNF), a novel embedding method from graph-structured data to real vectors based on a family of graph neural networks. The embedding naturally deals with graph isomorphism and preserves, in probability, the metric structure of graph domain. In addition to being an explicit embedding method, it also allows to efficiently and effectively approximate graph metric distances (as well as complete kernel functions); a criterion to select the embedding dimension trading off the approximation accuracy with the computational cost is also provided. Derived GRNF can be used within traditional processing methods or as input layer of a larger graph neural network. The theoretical guarantees that accompany GRNF ensure that the considered graph distance is metric, hence allowing to distinguish any pair of non-isomorphic graphs.

READ FULL TEXT

Distance-Preserving Graph Embeddings from Random Neural Features

Conditional Graph Neural Processes: A Functional Autoencoder Approach

Graph Neural Distance Metric Learning with Graph-Bert

Improving Expressivity of Graph Neural Networks

Learning the Geodesic Embedding with Graph Neural Networks

Embedding Signals on Knowledge Graphs with Unbalanced Diffusion Earth Mover's Distance

Accurate and Fast Retrieval for Complex Non-metric Data via Neighborhood Graphs

Expectation-Complete Graph Representations with Homomorphisms

Distance-Preserving Graph Embeddings from Random Neural Features

Related Research

Conditional Graph Neural Processes: A Functional Autoencoder Approach

Graph Neural Distance Metric Learning with Graph-Bert

Improving Expressivity of Graph Neural Networks

Learning the Geodesic Embedding with Graph Neural Networks

Embedding Signals on Knowledge Graphs with Unbalanced Diffusion Earth Mover's Distance

Accurate and Fast Retrieval for Complex Non-metric Data via Neighborhood Graphs

Expectation-Complete Graph Representations with Homomorphisms