Dirichlet Graph Variational Autoencoder
Lists
Only you see this
Abstract
Graph Neural Networks (GNNs) and Variational Autoencoders (VAEs) have been widely used in modeling and generating graphs with latent factors. However, there is no clear explanation of what these latent factors are and why they perform well. In this work, we present Dirichlet Graph Variational Autoencoder (DGVAE) with graph cluster memberships as latent factors. Our study connects VAEs based graph generation and balanced graph cut, and provides a new way to understand and improve the internal mechanism of VAEs based graph generation. Specifically, we first interpret the reconstruction term of DGVAE as balanced graph cut in a principled way. Furthermore, motivated by the low pass characteristics in balanced graph cut, we propose a new variant of GNN named Heatts to encode the input graph into cluster memberships. Heatts utilizes the Taylor series for fast computation of heat kernels and has better low pass characteristics than Graph Convolutional Networks (GCN). Through experiments on graph generation and graph clustering, we demonstrate the effectiveness of our proposed framework.
BibTeX
@inproceedings{li2020dirichlet,
author = {Jia Li and Tomas Yu and Jiajin Li and Honglei Zhang and Kangfei Zhao and Yu Rong and Hong Cheng and Junzhou Huang},
title = {{Dirichlet Graph Variational Autoencoder}},
booktitle = {Advances in Neural Information Processing Systems},
year = {2020},
url = {https://openreview.net/forum?id=IAmQCnaAVT},
}| Variational Graph Auto-Encoders | arXiv · 2016 |
| Directed Graph Generation with Heat Kernels | TMLR · 2025 |
| Multiresolution equivariant graph variational autoencoder | Mach. Learn. Sci. Technol · 2023 |
| Variational Graph Embedding and Clustering with Laplacian Eigenmaps | IJCAI · 2019 |
| Graph Community Augmentation with GMM-Based Modeling in Latent Space | arXiv · 2024 |
| xVGAE: A Hierarchical Variational Graph Autoencoder for Exchangeable Graphs | arXiv · 2026 |