Adaptive Model Compression of Ensembles for Evolving Data Streams Forecasting - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Adaptive Model Compression of Ensembles for Evolving Data Streams Forecasting

Résumé

Ensemble methods combining several models have shown superior predictive performance in data streams forecasting compared to individual models. Besides, they can cope with evolving data streams and concept drift as they allow adaptation. However, ensembles are renowned for their complexity and computational costs which makes them unsuitable in cases where both resources and time are limited such as IoT applications. In this paper, we propose to use model compression in the streaming setting in order to overcome the aforementioned drawbacks. We show that compressing a highly performing dynamic ensemble into an individual model leads to better predictive performance when compared to an individual learner while significantly reducing computational costs. We conduct an extensive experimental study on both real and synthetic time series to measure the impact of compression on both predictive performance and computational cost.
Fichier non déposé

Dates et versions

hal-04468401 , version 1 (20-02-2024)

Identifiants

Citer

Dihia Boulegane, Vitor Cerquiera, Albert Bifet. Adaptive Model Compression of Ensembles for Evolving Data Streams Forecasting. International Joint Conference on Neural Networks, IJCNN 2022, Padua, Italy, July 18-23, 2022, Jul 2022, Padua, Italy. pp.1--8, ⟨10.1109/IJCNN55064.2022.9892811⟩. ⟨hal-04468401⟩
6 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More