Enregistré dans:
Détails bibliographiques
Auteurs principaux: Cortés, Adrien, Rehm, Rémi, Letzelter, Victor
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2506.05515
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918121931014144
author Cortés, Adrien
Rehm, Rémi
Letzelter, Victor
author_facet Cortés, Adrien
Rehm, Rémi
Letzelter, Victor
contents We introduce TimeMCL, a method leveraging the Multiple Choice Learning (MCL) paradigm to forecast multiple plausible time series futures. Our approach employs a neural network with multiple heads and utilizes the Winner-Takes-All (WTA) loss to promote diversity among predictions. MCL has recently gained attention due to its simplicity and ability to address ill-posed and ambiguous tasks. We propose an adaptation of this framework for time-series forecasting, presenting it as an efficient method to predict diverse futures, which we relate to its implicit quantization objective. We provide insights into our approach using synthetic data and evaluate it on real-world time series, demonstrating its promising performance at a light computational cost.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05515
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Winner-takes-all for Multivariate Probabilistic Time Series Forecasting
Cortés, Adrien
Rehm, Rémi
Letzelter, Victor
Machine Learning
Artificial Intelligence
We introduce TimeMCL, a method leveraging the Multiple Choice Learning (MCL) paradigm to forecast multiple plausible time series futures. Our approach employs a neural network with multiple heads and utilizes the Winner-Takes-All (WTA) loss to promote diversity among predictions. MCL has recently gained attention due to its simplicity and ability to address ill-posed and ambiguous tasks. We propose an adaptation of this framework for time-series forecasting, presenting it as an efficient method to predict diverse futures, which we relate to its implicit quantization objective. We provide insights into our approach using synthetic data and evaluate it on real-world time series, demonstrating its promising performance at a light computational cost.
title Winner-takes-all for Multivariate Probabilistic Time Series Forecasting
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.05515