Learning emergent PDEs in a learned emergent space

Abstract

We extract data-driven, intrinsic spatial coordinates from observations of the dynamics of large systems of coupled heterogeneous agents. These coordinates then serve as an emergent space in which to learn predictive models in the form of partial differential equations (PDEs) for the collective description of the coupled-agent system. They play the role of the independent spatial variables in this PDE (as opposed to the dependent, possibly also data-driven, state variables). This leads to an alternative description of the dynamics, local in these emergent coordinates, thus facilitating an alternative modeling path for complex coupled-agent systems. We illustrate this approach on a system where each agent is a limit cycle oscillator (a so-called Stuart-Landau oscillator); the agents are heterogeneous (they each have a different intrinsic frequency $\omega$) and are coupled through the ensemble average of their respective variables. After fast initial transients, we show that the collective dynamics on a slow manifold can be approximated through a learned model based on local "spatial" partial derivatives in the emergent coordinates. The model is then used for prediction in time, as well as to capture collective bifurcations when system parameters vary. The proposed approach thus integrates the automatic, data-driven extraction of emergent space coordinates parametrizing the agent dynamics, with machine-learning assisted identification of an "emergent PDE" description of the dynamics in this parametrization.

References in corpus (11)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• DeepXDE: A deep learning library for solving differential equations
• Input Convex Neural Networks
• Long-time predictive modeling of nonlinear dynamical systems using neural networks
• Deep learning to discover and predict dynamics on an inertial manifold
• Reduced models in chemical kinetics via nonlinear data-mining
• Robust Learning with Jacobian Regularization
• Coarse-graining the Dynamics of a Driven Interface in the Presence of Mobile Impurities: Effective Description via Diffusion Maps
• An Emergent Space for Distributed Data with Hidden Internal Order through Manifold Learning
• Learning Contracting Vector Fields For Stable Imitation Learning

Cited by in corpus (2)

• Statistical Analysis of Tipping Pathways in Agent-Based Models
• Particles to Partial Differential Equations Parsimoniously