Continuous Graph Neural Networks

This paper builds on the connection between graph neural networks and traditional dynamical systems. We propose continuous graph neural networks (CGNN), which generalise existing graph neural networks with discrete dynamics in that they can be viewed as a specific discretisation scheme. The key idea is how to characterise the continuous dynamics of node representations, i.e. the derivatives of node representations, w.r.t. time. Inspired by existing diffusion-based methods on graphs (e.g. PageRank and epidemic models on social networks), we define the derivatives as a combination of the current node representations, the representations of neighbors, and the initial values of the nodes. We propose and analyse two possible dynamics on graphs---including each dimension of node representations (a.k.a. the feature channel) change independently or interact with each other---both with theoretical justification. The proposed continuous graph neural networks are robust to over-smoothing and hence allow us to build deeper networks, which in turn are able to capture the long-range dependencies between nodes. Experimental results on the task of node classification demonstrate the effectiveness of our proposed approach over competitive baselines.

[1]  David Duvenaud,et al.  Latent ODEs for Irregularly-Sampled Time Series , 2019, ArXiv.

[2]  Anderson R. Tavares,et al.  Discrete and Continuous Deep Residual Learning Over Graphs , 2019, ICAART.

[3]  M. Hirsch,et al.  Differential Equations, Dynamical Systems, and an Introduction to Chaos , 2003 .

[4]  Stephan Günnemann,et al.  Diffusion Improves Graph Learning , 2019, NeurIPS.

[5]  Megha Nawhal,et al.  Continuous Graph Flow for Flexible Density Estimation , 2019, ArXiv.

[6]  Yixin Chen,et al.  Link Prediction Based on Graph Neural Networks , 2018, NeurIPS.

[7]  Takanori Maehara,et al.  Revisiting Graph Neural Networks: All We Have is Low-Pass Filters , 2019, ArXiv.

[8]  Kilian Q. Weinberger,et al.  Simplifying Graph Convolutional Networks , 2019, ICML.

[9]  Doina Precup,et al.  Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks , 2019, NeurIPS.

[10]  Moustapha Cissé,et al.  Parseval Networks: Improving Robustness to Adversarial Examples , 2017, ICML.

[11]  Taiji Suzuki,et al.  Graph Neural Networks Exponentially Lose Expressive Power for Node Classification , 2019, ICLR.

[12]  Ameet Talwalkar,et al.  Massively Parallel Hyperparameter Tuning , 2018, ArXiv.

[13]  Yee Whye Teh,et al.  Augmented Neural ODEs , 2019, NeurIPS.

[14]  Guillaume Lample,et al.  Word Translation Without Parallel Data , 2017, ICLR.

[15]  Pietro Liò,et al.  Graph Attention Networks , 2017, ICLR.

[16]  Yuan Luo,et al.  Graph Convolutional Networks for Text Classification , 2018, AAAI.

[17]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[18]  Rajeev Motwani,et al.  The PageRank Citation Ranking : Bringing Order to the Web , 1999, WWW 1999.

[19]  Diego Marcheggiani,et al.  Encoding Sentences with Graph Convolutional Networks for Semantic Role Labeling , 2017, EMNLP.

[20]  Yoshua Bengio,et al.  GMNN: Graph Markov Neural Networks , 2019, ICML.

[21]  Jian-Yun Nie,et al.  RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space , 2018, ICLR.

[22]  Juntang Zhuang,et al.  Ordinary differential equations on graph networks , 2019 .

[23]  Stephan Günnemann,et al.  Predict then Propagate: Combining neural networks with personalized pagerank for classification on graphs , 2018, ICLR 2018.

[24]  Zhiyuan Liu,et al.  Graph Neural Networks: A Review of Methods and Applications , 2018, AI Open.

[25]  Bernhard Schölkopf,et al.  Learning with Local and Global Consistency , 2003, NIPS.

[26]  U. Feige,et al.  Spectral Graph Theory , 2015 .

[27]  Hajime Asama,et al.  Graph Neural Ordinary Differential Equations , 2019, ArXiv.

[28]  Xiao-Ming Wu,et al.  Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning , 2018, AAAI.

[29]  Peter Benner,et al.  Solution formulas for differential Sylvester and Lyapunov equations , 2018, Calcolo.

[30]  Xavier Bresson,et al.  Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering , 2016, NIPS.

[31]  Rose Yu,et al.  Understanding the Representation Power of Graph Neural Networks in Learning Graph Topology , 2019, NeurIPS.

[32]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[33]  Samuel S. Schoenholz,et al.  Neural Message Passing for Quantum Chemistry , 2017, ICML.

[34]  Ken-ichi Kawarabayashi,et al.  Representation Learning on Graphs with Jumping Knowledge Networks , 2018, ICML.

[35]  Ruslan Salakhutdinov,et al.  Revisiting Semi-Supervised Learning with Graph Embeddings , 2016, ICML.

[36]  David Duvenaud,et al.  Neural Ordinary Differential Equations , 2018, NeurIPS.

[37]  Lili Meng,et al.  Continuous Graph Flow , 2019 .

[38]  S. Sieniutycz,et al.  Optimal Control: An Introduction , 2001 .

[39]  M. L. Chambers.  The Mathematical Theory of Optimal Processes , 1965 .

[40]  W. O. Kermack,et al.  A contribution to the mathematical theory of epidemics , 1927 .

[41]  Qing Wang,et al.  DFNets: Spectral CNNs for Graphs with Feedback-Looped Filters , 2019, NeurIPS.