Generative machine learning methods for multivariate ensemble post-processing

Abstract

Ensemble weather forecasts based on multiple runs of numerical weather prediction models typically show systematic errors and require post-processing to obtain reliable forecasts. Accurately modeling multivariate dependencies is crucial in many practical applications, and various approaches to multivariate post-processing have been proposed where ensemble predictions are first post-processed separately in each margin and multivariate dependencies are then restored via copulas. These two-step methods share common key limitations, in particular the difficulty to include additional predictors in modeling the dependencies. We propose a novel multivariate post-processing method based on generative machine learning to address these challenges. In this new class of nonparametric data-driven distributional regression models, samples from the multivariate forecast distribution are directly obtained as output of a generative neural network. The generative model is trained by optimizing a proper scoring rule which measures the discrepancy between the generated and observed data, conditional on exogenous input variables. Our method does not require parametric assumptions on univariate distributions or multivariate dependencies and allows for incorporating arbitrary predictors. In two case studies on multivariate temperature and wind speed forecasting at weather stations over Germany, our generative model shows significant improvements over state-of-the-art methods and particularly improves the representation of spatial dependencies.

References in corpus (37)

• Adam: A Method for Stochastic Optimization
• Scikit-learn: Machine Learning in Python
• Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
• InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
• Skillful Precipitation Nowcasting using Deep Generative Models of Radar
• Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles
• Equivalence of distance-based and RKHS-based statistics in hypothesis testing
• Neural networks for post-processing ensemble weather forecasts
• Generative Moment Matching Networks
• Uncertainty Quantification in Complex Simulation Models Using Ensemble Copula Coupling
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Statistical Postprocessing for Weather Forecasts -- Review, Challenges and Avenues in a Big Data World
• A Generative Deep Learning Approach to Stochastic Downscaling of Precipitation Forecasts
• Stochastic Super-Resolution for Downscaling Time-Evolving Atmospheric Fields with a Generative Adversarial Network
• Training generative neural networks via Maximum Mean Discrepancy optimization
• Learning in Implicit Generative Models
• Improved Techniques for Training Score-Based Generative Models
• Multivariate probabilistic forecasting using Bayesian model averaging and copulas
• Log-normal distribution based EMOS models for probabilistic wind speed forecasting
• Comparison of nonhomogeneous regression models for probabilistic wind speed forecasting
• Ensemble model output statistics for wind vectors
• Spatial postprocessing of ensemble forecasts for temperature using nonhomogeneous Gaussian regression
• Probabilistic wind speed forecasting on a grid based on ensemble model output statistics
• Mixture EMOS model for calibrating ensemble forecasts of wind speed
• A similarity-based implementation of the Schaake shuffle
• Similarity-based semi-local estimation of EMOS models
• Generation of scenarios from calibrated ensemble forecasts with a dual ensemble copula coupling approach
• Increasing the accuracy and resolution of precipitation forecasts using deep generative models
• Joint probabilistic forecasting of wind speed and temperature using Bayesian model averaging
• Comparison of multivariate post-processing methods using global ECMWF ensemble forecasts
• Bivariate Gaussian models for wind vectors in a distributional regression framework
• Truncated generalized extreme value distribution based EMOS model for calibration of wind speed ensemble forecasts
• Probabilistic Forecasting with Generative Networks via Scoring Rule Minimization
• Probabilistic multivariate electricity price forecasting using implicit generative ensemble post-processing
• Evaluating forecasts for high-impact events using transformed kernel scores
• Convolutional autoencoders for spatially-informed ensemble post-processing
• Multivariate Forecasting Evaluation: On Sensitive and Strictly Proper Scoring Rules

Cited by in corpus (7)

• A Gentle Introduction and Tutorial on Deep Generative Models in Transportation Research
• Postprocessing of point predictions for probabilistic forecasting of day-ahead electricity prices: The benefits of using isotonic distributional regression
• Uncertainty Quantification of Wind Gust Predictions in the Northeast United States: An Evidential Neural Network and Explainable Artificial Intelligence Approach
• Machine learning-based probabilistic forecasting of solar irradiance in Chile
• Generative Machine Learning for Multivariate Angular Simulation
• A Composite-Loss Graph Neural Network for the Multivariate Post-Processing of Ensemble Weather Forecasts
• Enhancing multivariate post-processed visibility predictions utilizing CAMS forecasts