Capturing Graphs with Hypo-Elliptic Diffusions

by   Csaba Tóth, et al.

Convolutional layers within graph neural networks operate by aggregating information about local neighbourhood structures; one common way to encode such substructures is through random walks. The distribution of these random walks evolves according to a diffusion equation defined using the graph Laplacian. We extend this approach by leveraging classic mathematical results about hypo-elliptic diffusions. This results in a novel tensor-valued graph operator, which we call the hypo-elliptic graph Laplacian. We provide theoretical guarantees and efficient low-rank approximation algorithms. In particular, this gives a structured approach to capture long-range dependencies on graphs that is robust to pooling. Besides the attractive theoretical properties, our experiments show that this method competes with graph transformers on datasets requiring long-range reasoning but scales only linearly in the number of edges as opposed to quadratically in nodes.


page 1

page 2

page 3

page 4


MGNNI: Multiscale Graph Neural Networks with Implicit Layers

Recently, implicit graph neural networks (GNNs) have been proposed to ca...

Are Graph Convolutional Networks Fully Exploiting Graph Structure?

Graph Convolutional Networks (GCNs) generalize the idea of deep convolut...

Implicit Graph Neural Networks

Graph Neural Networks (GNNs) are widely used deep learning models that l...

Random Walks on Simplicial Complexes and the normalized Hodge Laplacian

Modeling complex systems and data with graphs has been a mainstay of the...

Graph Kernel Attention Transformers

We introduce a new class of graph neural networks (GNNs), by combining s...

Compatibility, embedding and regularization of non-local random walks on graphs

Several variants of the graph Laplacian have been introduced to model no...

Quasi-Monte Carlo Graph Random Features

We present a novel mechanism to improve the accuracy of the recently-int...

Please sign up or login with your details

Forgot password? Click here to reset