Graph Autoencoders with Deconvolutional Networks

TL;DR

This paper introduces Graph Deconvolutional Networks (GDNs) as the inverse of GCNs to reconstruct original graph signals from smoothed representations, enhancing graph autoencoder capabilities for various tasks.

Contribution

It proposes a novel GDN architecture combining spectral inverse filters and wavelet de-noising, enabling effective graph signal reconstruction in autoencoders.

Findings

GDN effectively reconstructs original graph signals from smoothed representations.

The proposed autoencoder improves performance on graph-level tasks.

GDN enhances graph generation and social recommendation accuracy.

Abstract

Recent studies have indicated that Graph Convolutional Networks (GCNs) act as a \emph{low pass} filter in spectral domain and encode smoothed node representations. In this paper, we consider their opposite, namely Graph Deconvolutional Networks (GDNs) that reconstruct graph signals from smoothed node representations. We motivate the design of Graph Deconvolutional Networks via a combination of inverse filters in spectral domain and de-noising layers in wavelet domain, as the inverse operation results in a \emph{high pass} filter and may amplify the noise. Based on the proposed GDN, we further propose a graph autoencoder framework that first encodes smoothed graph representations with GCN and then decodes accurate graph signals with GDN. We demonstrate the effectiveness of the proposed method on several tasks including unsupervised graph-level representation , social recommendation and graph generation