Mixed moving average field guided learning for spatio-temporal data

2023/01/02 by Curato, Imma Valentina, Furat, Orkun, Proietti, Lorenzo +1
#60E07 #60E15 #60G25 #60G60 #62C10 #FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Statistics Theory (math.ST)

paper · doi:10.48550/arxiv.2301.00736

Abstract

Influenced mixed moving average fields are a versatile modeling class for spatio-temporal data. However, their predictive distribution is not generally known. Under this modeling assumption, we define a novel spatio-temporal embedding and a theory-guided machine learning approach that employs a generalized Bayesian algorithm to make ensemble forecasts. We use Lipschitz predictors and determine fixed-time and any-time PAC Bayesian bounds in the batch learning setting. Performing causal forecast is a highlight of our methodology as its potential application to data with spatial and temporal short and long-range dependence. We then test the performance of our learning methodology by using linear predictors and data sets simulated from a spatio-temporal Ornstein-Uhlenbeck process.

Related

• Dynamic Bayesian Predictive Stacking via Markovian Spatiotemporal Propagation
• Beyond Fixed Variables: Expanding-variate Time Series Forecasting via Flat Scheme and Spatio-temporal Focal Learning
• Multi-Modality Spatio-Temporal Forecasting via Self-Supervised Learning
• Machine Learning for Spatiotemporal Sequence Forecasting: A Survey
• Deciphering Spatio-Temporal Graph Forecasting: A Causal Lens and Treatment