Deep learning of contagion dynamics on complex networks

Abstract

Forecasting the evolution of contagion dynamics is still an open problem to which mechanistic models only offer a partial answer. To remain mathematically or computationally tractable, these models must rely on simplifying assumptions, thereby limiting the quantitative accuracy of their predictions and the complexity of the dynamics they can model. Here, we propose a complementary approach based on deep learning where the effective local mechanisms governing a dynamic on a network are learned from time series data. Our graph neural network architecture makes very few assumptions about the dynamics, and we demonstrate its accuracy using different contagion dynamics of increasing complexity. By allowing simulations on arbitrary network structures, our approach makes it possible to explore the properties of the learned dynamics beyond the training data. Finally, we illustrate the applicability of our approach using real data of the COVID-19 outbreak in Spain. Our results demonstrate how deep learning offers a new and complementary perspective to build effective models of contagion dynamics on networks.

References in corpus (30)

• Adam: A Method for Stochastic Optimization
• DeepWalk: Online Learning of Social Representations
• Semi-Supervised Classification with Graph Convolutional Networks
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Inductive Representation Learning on Large Graphs
• Epidemic processes in complex networks
• Modeling polypharmacy side effects with graph convolutional networks
• Deep learning for universal linear embeddings of nonlinear dynamics
• Reaction-diffusion processes and metapopulation models in heterogeneous networks
• Simplicial models of social contagion
• What's in a crowd? Analysis of face-to-face behavioral networks
• Modeling the Worldwide Spread of Pandemic Influenza: Baseline Case and Containment Interventions
• Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data
• Contact patterns in a high school: a comparison between data collected using wearable sensors, contact diaries and friendship surveys
• How Powerful are Graph Neural Networks?
• Foundations and modelling of dynamic networks using Dynamic Graph Neural Networks: A survey
• Can co-location be used as a proxy for face-to-face contacts?
• Principal Neighbourhood Aggregation for Graph Nets
• Scale-free Networks Well Done
• Examining COVID-19 Forecasting using Spatio-Temporal Graph Neural Networks
• A framework for epidemic spreading in multiplex networks of metapopulations
• Discovery of Physics from Data: Universal Laws and Discrepancies
• Attention Interpretability Across NLP Tasks
• Interacting contagions are indistinguishable from social reinforcement
• A General Deep Learning Framework for Network Reconstruction and Dynamics Learning
• Koopman Operator and its Approximations for Systems with Symmetries
• Bayesian Inference of Spreading Processes on Networks
• A machine learning approach to predicting dynamical observables from network structure
• Combining Graph Neural Networks and Spatio-temporal Disease Models to Predict COVID-19 Cases in Germany
• Detecting structural perturbations from time series with deep learning

Cited by in corpus (9)

• Distinguishing simple and complex contagion processes on networks
• Epidemic Source Detection in Contact Tracing Networks: Epidemic Centrality in Graphs and Message-Passing Algorithms
• A Systematic Literature Review of Spatio-Temporal Graph Neural Network Models for Time Series Forecasting and Classification
• Nonlinear bias toward complex contagion in uncertain transmission settings
• Privacy-Preserving Individual-Level COVID-19 Infection Prediction via Federated Graph Learning
• Duality between predictability and reconstructability in complex systems
• Predicting Long-term Dynamics of Complex Networks via Identifying Skeleton in Hyperbolic Space
• Learning Continuous Network Emerging Dynamics from Scarce Observations via Data-Adaptive Stochastic Processes
• Reconstructing networks from simple and complex contagions