On Generalization of Graph Autoencoders with Adversarial Training
Abstract
Adversarial training is an approach for increasing model's resilience against adversarial perturbations. Such approaches have been demonstrated to result in models with feature representations that generalize better. However, limited works have been done on adversarial training of models on graph data. In this paper, we raise such a question { does adversarial training improve the generalization of graph representations. We formulate L2 and L1 versions of adversarial training in two powerful node embedding methods: graph autoencoder (GAE) and variational graph autoencoder (VGAE). We conduct extensive experiments on three main applications, i.e. link prediction, node clustering, graph anomaly detection of GAE and VGAE, and demonstrate that both L2 and L1 adversarial training boost the generalization of GAE and VGAE.
Cite
Text
Huang et al. "On Generalization of Graph Autoencoders with Adversarial Training." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021. doi:10.1007/978-3-030-86520-7_23Markdown
[Huang et al. "On Generalization of Graph Autoencoders with Adversarial Training." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021.](https://mlanthology.org/ecmlpkdd/2021/huang2021ecmlpkdd-generalization/) doi:10.1007/978-3-030-86520-7_23BibTeX
@inproceedings{huang2021ecmlpkdd-generalization,
title = {{On Generalization of Graph Autoencoders with Adversarial Training}},
author = {Huang, Tianjin and Pei, Yulong and Menkovski, Vlado and Pechenizkiy, Mykola},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2021},
pages = {367-382},
doi = {10.1007/978-3-030-86520-7_23},
url = {https://mlanthology.org/ecmlpkdd/2021/huang2021ecmlpkdd-generalization/}
}