A mixture of experts model for predicting persistent weather patterns

Áreas de investigación:

• Methodology - Machine learning

Año:

2018

Tipo de publicación:

Artículo en conferencia

Palabras clave:

mixture of experts, persistence model, dynamic systems, ordinal classification, ordinal regression, autoregressive models, neural networks, low-visibility

Autores:

• Pérez-Ortiz, María
• Gutiérrez, Pedro Antonio
• Tino, Peter
• Casanova-Mateo, Carlos
• Salcedo-Sanz, Sancho

Título del libro:

Proceedings of the 2018 IEEE International Joint Conference on Neural Networks (IJCNN 2018)

Páginas:

5714-5721

Organización:

Rio (Brazil)

Mes:

8th-13th July

ISBN:

978-1-5090-6014-6

BibTex:

@conference{IJCNN_2018, author = "Mar{\'i}a P{\'e}rez-Ortiz and Pedro Antonio Guti{\'e}rrez and Peter Tino and Carlos Casanova-Mateo and Sancho Salcedo-Sanz", abstract = "Weather and atmospheric patterns are often persistent. The simplest weather forecasting method is the so-called persistence model, which assumes that the future state of a system will be similar (or equal) to the present state. Machine learning (ML) models are widely used in different weather forecasting applications, however, these models need to be compared to the persistence model to analyse whether ML provides a competitive solution to the problem at hand. In this paper we devise a new model for predicting low-visibility in airports using the concepts of mixture of experts. Visibility level is coded as two different ordered categorical variables (cloud height and runway visual height). The underlying system in this application is stagnant approximately in 90% of the cases and standard ML models fail to improve on the performance of the persistence model. Because of this, instead of trying to simply beat the persistence model using ML we use this persistence as a baseline and learn a ordinal neural network model that refines its results by focusing on learning weather fluctuations. The results show that the proposal outperforms persistence and other ordinal autoregressive models, especially for longer time horizon predictions and for the runway visual height variable.", booktitle = "Proceedings of the 2018 IEEE International Joint Conference on Neural Networks (IJCNN 2018)", doi = "10.1109/IJCNN.2018.8489179", isbn = "978-1-5090-6014-6", keywords = "mixture of experts, persistence model, dynamic systems, ordinal classification, ordinal regression, autoregressive models, neural networks, low-visibility", month = "8th-13th July", organization = "Rio (Brazil)", pages = "5714--5721", publisher = "IEEE Press", title = "{A} mixture of experts model for predicting persistent weather patterns", url = "doi.org/10.1109/IJCNN.2018.8489179", year = "2018", }

Abstract:

Weather and atmospheric patterns are often persistent. The simplest weather forecasting method is the so-called persistence model, which assumes that the future state of a system will be similar (or equal) to the present state. Machine learning (ML) models are widely used in different weather forecasting applications, however, these models need to be compared to the persistence model to analyse whether ML provides a competitive solution to the problem at hand. In this paper we devise a new model for predicting low-visibility in airports using the concepts of mixture of experts. Visibility level is coded as two different ordered categorical variables (cloud height and runway visual height). The underlying system in this application is stagnant approximately in 90% of the cases and standard ML models fail to improve on the performance of the persistence model. Because of this, instead of trying to simply beat the persistence model using ML we use this persistence as a baseline and learn a ordinal neural network model that refines its results by focusing on learning weather fluctuations. The results show that the proposal outperforms persistence and other ordinal autoregressive models, especially for longer time horizon predictions and for the runway visual height variable.