Abstract:Predictive linear and nonlinear models based on kernel machines or deep neural networks have been used to discover dependencies among time series. This paper proposes an efficient nonlinear modeling approach for multiple time series, with a complexity comparable to linear vector autoregressive (VAR) models while still incorporating nonlinear interactions among different time-series variables. The modeling assumption is that the set of time series is generated in two steps: first, a linear VAR process in a latent space, and second, a set of invertible and Lipschitz continuous nonlinear mappings that are applied per sensor, that is, a component-wise mapping from each latent variable to a variable in the measurement space. The VAR coefficient identification provides a topology representation of the dependencies among the aforementioned variables. The proposed approach models each component-wise nonlinearity using an invertible neural network and imposes sparsity on the VAR coefficients to reflect the parsimonious dependencies usually found in real applications. To efficiently solve the formulated optimization problems, a custom algorithm is devised combining proximal gradient descent, stochastic primal-dual updates, and projection to enforce the corresponding constraints. Experimental results on both synthetic and real data sets show that the proposed algorithm improves the identification of the support of the VAR coefficients in a parsimonious manner while also improving the time-series prediction, as compared to the current state-of-the-art methods.

A Decentralized Approach for Nonlinear Prediction of Time Series Data in Sensor Networks

Neural Network Emulation of Spatio-temporal Data Using Linear and Nonlinear Dimensionality Reduction

Adaptation and Learning over Networks for Nonlinear System Modeling

Non-negative distributed regression for data inference in wireless sensor networks

Distributed Self-Monitoring Sensor Networks Via Markov Switching Dynamic Linear Models

Sparse particle filtering for modeling space-time dynamics in distributed sensor networks

Adaptive Virtual Sensors Using SNPER for the Localized Construction and Elastic Net Regularization in Nonlinear Processes

Distributed Prediction Via Adaptive Hammerstein Filter over Networked Systems

Efficient Interpretable Nonlinear Modeling for Multiple Time Series

Learn to Sense: a Meta-learning Based Distributed Sensing Algorithm in Wireless Sensor Networks

Distributed Extreme Learning Machine for Nonlinear Learning over Network

Online learning algorithm for time series forecasting suitable for low cost wireless sensor networks nodes

Lightweight Self-Adapting Linear Prediction Algorithms for Wireless Sensor Networks

Hierarchical Event-triggered Learning for Cyclically Excited Systems with Application to Wireless Sensor Networks

A Distributed Training Method for Sparse Kernel Machine over WSN

Distributed Noise Covariance Matrices Estimation in Sensor Networks

Energy-Balanced Distributed Sparse Kernel Machine In Wireless Sensor Network

Decentralized event-triggered estimation of nonlinear systems

Modeling and Analysis of Data Prediction Technique Based on Linear Regression Model (DP-LRM) for Cluster-Based Sensor Networks

A Distributed Learning Method for ℓ 1 -Regularized Kernel Machine over Wireless Sensor Networks

Modeling Randomly Observed Spatiotemporal Dynamical Systems