Graph Neural Networks with a Distribution of Parametrized Graphs
Not on the map. Why?
Abstract (arXiv)
Traditionally, graph neural networks have been trained using a single observed graph. However, the observed graph represents only one possible realization. In many applications, the graph may encounter uncertainties, such as having erroneous or missing edges, as well as edge weights that provide little informative value. To address these challenges and capture additional information previously absent in the observed graph, we introduce latent variables to parameterize and generate multiple graphs. We obtain the maximum likelihood estimate of the network parameters in an Expectation-Maximization (EM) framework based on the multiple graphs. Specifically, we iteratively determine the distribution of the graphs using a Markov Chain Monte Carlo (MCMC) method, incorporating the principles of PAC-Bayesian theory. Numerical experiments demonstrate improvements in performance against baseline models on node classification for heterogeneous graphs and graph regression on chemistry datasets.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 33 of its 47 references that are in the map. First posted on arXiv 25 Oct 2023.
Builds on
33 of 47 references.
References · 33
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
- 2017Graph Attention Networkscited by 8,854
- 2018How Powerful are Graph Neural Networks?cited by 4,355
- 2017Neural Message Passing for Quantum Chemistrycited by 3,460
- 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,085
- 2018Graph Convolutional Neural Networks for Web-Scale Recommender Systemscited by 1,133
- 2020Simple and Deep Graph Convolutional Networkscited by 857
- 2021How Attentive are Graph Attention Networks?cited by 818
- 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
- 2019DropEdge: Towards Deep Graph Convolutional Networks on Node Classificationcited by 679
- 2015Train faster, generalize better: Stability of stochastic gradient descentcited by 547
- 2019Graph Transformer Networkscited by 520
21 more
- 2021Understanding over-squashing and bottlenecks on graphs via curvaturecited by 471
- 2016Molecular Graph Convolutions: Moving Beyond Fingerprintscited by 238
- 2021Are we really making much progress? Revisiting, benchmarking, and refining heterogeneous graph neural networkscited by 176
- 2022Revisiting Heterophily For Graph Neural Networkscited by 169
- 2019A Primer on PAC-Bayesian Learningcited by 130
- 2018Bayesian graph convolutional neural networks for semi-supervised classificationcited by 88
- 2022Simple and Efficient Heterogeneous Graph Neural Networkcited by 71
- 2022Towards Robust Graph Neural Networks for Noisy Graphs with Sparse Labelscited by 63
- 2021DGL-LifeSci: An Open-Source Toolkit for Deep Learning on Graphs in Life Sciencecited by 38
- 2021Modeling Scale-free Graphs with Hyperbolic Geometry for Knowledge-aware Recommendationcited by 38
- 2022ACMP: Allen-Cahn Message Passing with Attractive and Repulsive Forces for Graph Neural Networkscited by 31
- 2020Heterogeneous Molecular Graph Neural Networks for Predicting Molecule Propertiescited by 30
- 2023Graph Neural Convection-Diffusion with Heterophilycited by 26
- 2021Curvature Graph Neural Networkcited by 24
- 2022SGAT: Simplicial Graph Attention Networkcited by 16
- 2021CogDL: A Comprehensive Library for Graph Deep Learningcited by 12
- 2023Node Embedding from Neural Hamiltonian Orbits in Graph Neural Networkscited by 12
- 2023Leveraging Label Non-Uniformity for Node Classification in Graph Neural Networkscited by 11
- 2021Graph Signal Processing over a Probability Space of Shift Operatorscited by 5
- 2023On distributional graph signalscited by 3
- 2021Learning on heterogeneous graphs using high-order relationscited by 2
Built on by
Citing papers per year: 2025: 1.
1 paper.