Generalizing to New Physical Systems via Context-Informed Dynamics Model - Archive ouverte HAL Access content directly
Conference Papers Year :

Generalizing to New Physical Systems via Context-Informed Dynamics Model


Data-driven approaches to modeling physical systems fail to generalize to unseen systems that share the same general dynamics with the learning domain, but correspond to different physical contexts. We propose a new framework for this key problem, context-informed dynamics adaptation (CoDA), which takes into account the distributional shift across systems for fast and efficient adaptation to new dynamics. CoDA leverages multiple environments, each associated to a different dynamic, and learns to condition the dynamics model on contextual parameters, specific to each environment. The conditioning is performed via a hypernetwork, learned jointly with a context vector from observed data. The proposed formulation constrains the search hypothesis space to foster fast adaptation and better generalization across environments. We theoretically motivate our approach and show state-of-the-art generalization results on a set of nonlinear dynamics, representative of a variety of application domains. We also show, on these systems, that new system parameters can be inferred from context vectors with minimal supervision.
Fichier principal
Vignette du fichier
ICML2022_CoDA.pdf (2.31 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03547546 , version 1 (31-01-2022)
hal-03547546 , version 2 (16-06-2022)



Matthieu Kirchmeyer, Yuan Yin, Jérémie Donà, Nicolas Baskiotis, Alain Rakotomamonjy, et al.. Generalizing to New Physical Systems via Context-Informed Dynamics Model. International Conference on Machine Learning, Jul 2022, Baltimore, France. ⟨hal-03547546v2⟩
267 View
98 Download



Gmail Facebook Twitter LinkedIn More