Learning Dynamic Graph Embeddings with Neural Controlled Differential Equations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qin, Tiexin, Walker, Benjamin, Lyons, Terry, Yan, Hong, Li, Haoliang
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914068536754176
author Qin, Tiexin
Walker, Benjamin
Lyons, Terry
Yan, Hong
Li, Haoliang
author_facet Qin, Tiexin
Walker, Benjamin
Lyons, Terry
Yan, Hong
Li, Haoliang
contents This paper focuses on representation learning for dynamic graphs with temporal interactions. A fundamental issue is that both the graph structure and the nodes own their own dynamics, and their blending induces intractable complexity in the temporal evolution over graphs. Drawing inspiration from the recent progress of physical dynamic models in deep neural networks, we propose Graph Neural Controlled Differential Equations (GN-CDEs), a continuous-time framework that jointly models node embeddings and structural dynamics by incorporating a graph enhanced neural network vector field with a time-varying graph path as the control signal. Our framework exhibits several desirable characteristics, including the ability to express dynamics on evolving graphs without piecewise integration, the capability to calibrate trajectories with subsequent data, and robustness to missing observations. Empirical evaluation on a range of dynamic graph representation learning tasks demonstrates the effectiveness of our proposed approach in capturing the complex dynamics of dynamic graphs.
format Preprint
id arxiv_https___arxiv_org_abs_2302_11354
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Learning Dynamic Graph Embeddings with Neural Controlled Differential Equations
Qin, Tiexin
Walker, Benjamin
Lyons, Terry
Yan, Hong
Li, Haoliang
Machine Learning
Artificial Intelligence
This paper focuses on representation learning for dynamic graphs with temporal interactions. A fundamental issue is that both the graph structure and the nodes own their own dynamics, and their blending induces intractable complexity in the temporal evolution over graphs. Drawing inspiration from the recent progress of physical dynamic models in deep neural networks, we propose Graph Neural Controlled Differential Equations (GN-CDEs), a continuous-time framework that jointly models node embeddings and structural dynamics by incorporating a graph enhanced neural network vector field with a time-varying graph path as the control signal. Our framework exhibits several desirable characteristics, including the ability to express dynamics on evolving graphs without piecewise integration, the capability to calibrate trajectories with subsequent data, and robustness to missing observations. Empirical evaluation on a range of dynamic graph representation learning tasks demonstrates the effectiveness of our proposed approach in capturing the complex dynamics of dynamic graphs.
title Learning Dynamic Graph Embeddings with Neural Controlled Differential Equations
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2302.11354