LEADS: Learning Dynamical Systems that Generalize Across Environments

TL;DR

LEADS is a framework that improves the generalization of dynamical system models across different environments by capturing shared and environment-specific dynamics, supported by theoretical and empirical results.

Contribution

It introduces a novel training formulation that leverages commonalities and discrepancies among environments to enhance model generalization, applicable to linear and nonlinear dynamics.

Findings

Theoretical reduction in sample complexity with LEADS.

Empirical improvement in generalization for nonlinear dynamics.

Effective exploitation of environment-dependent data.

Abstract

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. Code is available at https://github.com/yuan-yin/LEADS.