LEADS: Learning Dynamical Systems that Generalize Across Environments - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

LEADS: Learning Dynamical Systems that Generalize Across Environments

Résumé

When modeling dynamical systems from real-world data samples, the distribution of data often changes according to the environment in which they are captured, and the dynamics of the system itself vary from one environment to another. Generalizing across environments thus challenges the conventional frameworks. The classical settings suggest either considering data as i.i.d. and learning a single model to cover all situations or learning environment-specific models. Both are sub-optimal: the former disregards the discrepancies between environments leading to biased solutions, while the latter does not exploit their potential commonalities and is prone to scarcity problems. We propose LEADS, a novel framework that leverages the commonalities and discrepancies among known environments to improve model generalization. This is achieved with a tailored training formulation aiming at capturing common dynamics within a shared model while additional terms capture environment-specific dynamics. We ground our approach in theory, exhibiting a decrease in sample complexity with our approach and corroborate these results empirically, instantiating it for linear dynamics. Moreover, we concretize this framework for neural networks and evaluate it experimentally on representative families of nonlinear dynamics. We show that this new setting can exploit knowledge extracted from environment-dependent data and improves generalization for both known and novel environments.
Fichier principal
Vignette du fichier
neurips_2021.pdf (1.41 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03261055 , version 1 (15-06-2021)
hal-03261055 , version 2 (16-02-2022)

Identifiants

Citer

Yuan Yin, Ibrahim Ayed, Emmanuel de Bézenac, Nicolas Baskiotis, Patrick Gallinari. LEADS: Learning Dynamical Systems that Generalize Across Environments. The Thirty-Fifth Annual Conference on Neural Information Processing Systems (NeurIPS 2021), Dec 2021, Online, Unknown Region. ⟨hal-03261055v2⟩
169 Consultations
108 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More