We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Hyperbolic Variational Graph Neural Network for Modeling Dynamic Graphs

Abstract: Learning representations for graphs plays a critical role in a wide spectrum of downstream applications. In this paper, we summarize the limitations of the prior works in three folds: representation space, modeling dynamics and modeling uncertainty. To bridge this gap, we propose to learn dynamic graph representation in hyperbolic space, for the first time, which aims to infer stochastic node representations. Working with hyperbolic space, we present a novel Hyperbolic Variational Graph Neural Network, referred to as HVGNN. In particular, to model the dynamics, we introduce a Temporal GNN (TGNN) based on a theoretically grounded time encoding approach. To model the uncertainty, we devise a hyperbolic graph variational autoencoder built upon the proposed TGNN to generate stochastic node representations of hyperbolic normal distributions. Furthermore, we introduce a reparameterisable sampling algorithm for the hyperbolic normal distribution to enable the gradient-based learning of HVGNN. Extensive experiments show that HVGNN outperforms state-of-the-art baselines on real-world datasets.
Comments: 9 pages, Accepted as a regular paper in AAAI Conference on Artificial Intelligence 2021
Subjects: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Cite as: arXiv:2104.02228 [cs.LG]
  (or arXiv:2104.02228v1 [cs.LG] for this version)

Submission history

From: Li Sun [view email]
[v1] Tue, 6 Apr 2021 01:44:15 GMT (3687kb,D)

Link back to: arXiv, form interface, contact.