Machine Learning Link Inference of Noisy Delay-coupled Networks with Opto-Electronic Experimental Tests

Abstract

We devise a machine learning technique to solve the general problem of inferring network links that have time-delays. The goal is to do this purely from time-series data of the network nodal states. This task has applications in fields ranging from applied physics and engineering to neuroscience and biology. To achieve this, we first train a type of machine learning system known as reservoir computing to mimic the dynamics of the unknown network. We formulate and test a technique that uses the trained parameters of the reservoir system output layer to deduce an estimate of the unknown network structure. Our technique, by its nature, is non-invasive, but is motivated by the widely-used invasive network inference method whereby the responses to active perturbations applied to the network are observed and employed to infer network links (e.g., knocking down genes to infer gene regulatory networks). We test this technique on experimental and simulated data from delay-coupled opto-electronic oscillator networks. We show that the technique often yields very good results particularly if the system does not exhibit synchrony. We also find that the presence of dynamical noise can strikingly enhance the accuracy and ability of our technique, especially in networks that exhibit synchrony.

arXiv commentNew figures and subsections added

References in corpus (42)

• Statistical mechanics of complex networks
• Measuring Information Transfer
• Epidemic processes in complex networks
• Machine learning and the physical sciences
• Recent Advances in Physical Reservoir Computing: A Review
• Heterogeneity in oscillator networks: Are smaller worlds easier to synchronize?
• Optoelectronic Reservoir Computing
• Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data
• Symmetries, Cluster Synchronization, and Isolated Desynchronization in Complex Networks
• Attractor Reconstruction by Machine Learning
• Hybrid Forecasting of Chaotic Processes: Using Machine Learning in Conjunction with a Knowledge-Based Model
• Evaluating Link Prediction Methods
• High performance photonic reservoir computer based on a coherently driven passive cavity
• Functional modularity of background activities in normal and epileptic brain networks
• Causal Network Inference by Optimal Causation Entropy
• Perturbation Biology: inferring signaling networks in cellular systems
• Noise bridges dynamical correlation and topology in coupled oscillator networks
• Broadband chaos generated by an opto-electronic oscillator
• On Explaining the Surprising Success of Reservoir Computing Forecaster of Chaos? The Universal Machine Learning Dynamical System with Contrasts to VAR and DMD
• Robust Reconstruction of Complex Networks from Sparse Data
• Observability and Controllability of Nonlinear Networks: The Role of Symmetry
• Synchronizing distant nodes: a universal classification of networks
• Experimental observation of chimera and cluster states in a minimal globally coupled network
• Network reconstruction and community detection from dynamics
• Experimental Observations of Group Synchrony in a System of Chaotic Optoelectronic Oscillators
• Complex Dynamics and Synchronization of Delayed-Feedback Nonlinear Oscillators
• Breaking Synchrony by Heterogeneity in Complex Networks
• Robustness of Optimal Synchronization in Real Networks
• Reconstructing networks with unknown and heterogeneous errors
• Delayed Dynamical Systems: Networks, Chimeras and Reservoir Computing
• Separation of Chaotic Signals by Reservoir Computing
• Transition from phase to generalized synchronization in time-delay systems
• Using Machine Learning to Assess Short Term Causal Dependence and Infer Network Links
• Inferring network connectivity from event timing patterns
• Model reconstruction from temporal data for coupled oscillator networks
• Correlated fluctuations in strongly-coupled binary networks beyond equilibrium
• Machine learning with observers predicts complex spatiotemporal behavior
• Reconstructing dynamical networks via feature ranking
• Reverse engineering of complex dynamical networks in the presence of time-delayed interactions based on noisy time series
• Global generalized synchronization in networks of different time-delay systems
• Supervised Chaotic Source Separation by a Tank of Water
• Impact of network characteristics on network reconstruction

Cited by in corpus (8)

• Model-free prediction of multistability using echo state network
• Catch-22s of reservoir computing
• Time Shifts to Reduce the Size of Reservoir Computers
• Role of assortativity in predicting burst synchronization using echo state network
• Estimating the master stability function from the time series of one oscillator via reservoir computing
• Backpropagation on Dynamical Networks
• Learning Beyond Experience: Generalizing to Unseen State Space with Reservoir Computing
• Incorporating Coupling Knowledge into Echo State Networks for Learning Spatiotemporally Chaotic Dynamics