Deconvolutional Networks on Graph Data
arXiv:2110.15528
Abstract
In this paper, we consider an inverse problem in graph learning domain -- ``given the graph representations smoothed by Graph Convolutional Network (GCN), how can we reconstruct the input graph signal?" We propose Graph Deconvolutional Network (GDN) and motivate the design of GDN via a combination of inverse filters in spectral domain and de-noising layers in wavelet domain, as the inverse operation results in a high frequency amplifier and may amplify the noise. We demonstrate the effectiveness of the proposed method on several tasks including graph feature imputation and graph structure generation.
arXiv commentarXiv admin note: text overlap with arXiv:2012.11898
References in corpus (11)
- The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
- Simplifying Graph Convolutional Networks
- Graph Convolutional Matrix Completion
- Variational Graph Auto-Encoders
- DropEdge: Towards Deep Graph Convolutional Networks on Node Classification
- Learning Structural Node Embeddings Via Diffusion Wavelets
- Analyzing Inverse Problems with Invertible Neural Networks
- Subgraph Matching Kernels for Attributed Graphs
- Graph Convolutional Networks for Graphs Containing Missing Features
- Handling Missing Data with Graph Representation Learning
- Graph Wavelet Neural Network