Seasonal average temperature forecast with the AutoGluonTS modern autoML tool - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Seasonal average temperature forecast with the AutoGluonTS modern autoML tool

Abstract

Forecasting the temperature at a given place in a given future epoch is of evident huge interest. This is extremely hard unless future means "next days". In this paper, we focus on the problem of forecasting if a given future period (say, next summer) will be hotter than usually. More precisely, the objective is to say if the average daily temperature will be above normal, normal or below normal, where normal is defined on past observed values. For this task, most of the production tools are of the dynamic type (that is, physical, usually huge systems of partial differential equations) or statistical, or a combination of both. In the paper, we explore the use of modern Machine Learning tools and show that they can compete with those dedicated and always heavy and complex procedures, in spite of the fact that our results come from standard tools and hardware. We illustrate the results with data coming from a project where these questions are studied, concerning a large area in the Southern part of South America.
Fichier principal
Vignette du fichier
Rubino-MACLEAN23-CR.pdf (1.06 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Licence : Copyright

Dates and versions

hal-04384070 , version 1 (10-01-2024)

Licence

Copyright

Identifiers

  • HAL Id : hal-04384070 , version 1

Cite

Diego Kiedanski, Pablo Rodriguez-Bocca, Gerardo Rubino. Seasonal average temperature forecast with the AutoGluonTS modern autoML tool. MAChine Learning for EArth ObservatioN, European Conference on Machine Learning, Sep 2023, Torino, Italy. pp.1-10. ⟨hal-04384070⟩
32 View
15 Download

Share

Gmail Facebook X LinkedIn More