Data-driven discovery of emergent behaviors in collective dynamics
From MaRDI portal
Publication:2127368
DOI10.1016/J.PHYSD.2020.132542zbMATH Open1495.91098arXiv1912.11123OpenAlexW3025749473WikidataQ98205415 ScholiaQ98205415MaRDI QIDQ2127368FDOQ2127368
Authors: Yanyan Li 
Publication date: 20 April 2022
Published in: Physica D (Search for Journal in Brave)
Abstract: Particle- and agent-based systems are a ubiquitous modeling tool in many disciplines. We consider the fundamental problem of inferring interaction kernels from observations of agent-based dynamical systems given observations of trajectories, in particular for collective dynamical systems exhibiting emergent behaviors with complicated interaction kernels, in a nonparametric fashion, and for kernels which are parametrized by a single unknown parameter. We extend the estimators introduced in cite{PNASLU}, which are based on suitably regularized least squares estimators, to these larger classes of systems. We provide extensive numerical evidence that the estimators provide faithful approximations to the interaction kernels, and provide accurate predictions for trajectories started at new initial conditions, both throughout the ``training time interval in which the observations were made, and often much beyond. We demonstrate these features on prototypical systems displaying collective behaviors, ranging from opinion dynamics, flocking dynamics, self-propelling particle dynamics, synchronized oscillator dynamics, and a gravitational system. Our experiments also suggest that our estimated systems can display the same emergent behaviors of the observed systems, that occur at larger timescales than those used in the training data. Finally, in the case of families of systems governed by a parameterized family of interaction kernels, we introduce novel estimators that estimate the parameterized family of kernels, splitting it into a common interaction kernel and the action of parameters. We demonstrate this in the case of gravity, by learning both the ``common component and the dependency on mass, without any a priori knowledge of either one, from observations of planetary motions in our solar system.
Full work available at URL: https://arxiv.org/abs/1912.11123
Recommendations
- Nonparametric inference of interaction laws in systems of agents from trajectory data
- Inferring interaction rules from observations of evolutive systems I: The variational approach
- Learning theory for inferring interaction kernels in second-order interacting agent systems
- Nonlocal flocking dynamics: learning the fractional order of PDEs from particle simulations
- Kernel-based parameter estimation of dynamical systems with unknown observation functions
machine learningnonparametric inferencecollective dynamicsdata-driven methodagent-based dynamicsemergent behaviors
Cites Work
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- A distribution-free theory of nonparametric regression
- Title not available (Why is that?)
- Sparse dynamics for partial differential equations
- On the mathematics of emergence
- From Kuramoto to Crawford: Exploring the onset of synchronization in population of coupled oscillators
- A Primer of Swarm Equilibria
- Emergent Behavior in Flocks
- Approximation methods for supervised learning
- Heterophilious dynamics enhances consensus
- Hydrodynamic models of self-organized dynamics: derivation and existence theory
- A General Collision-Avoiding Flocking Framework
- Stability Analysis of Flock and Mill Rings for Second Order Models in Swarming
- Universal algorithms for learning theory. I: Piecewise constant functions.
- Double milling in self-propelled swarms from kinetic theory
- A discrete nonlinear and non-autonomous model of consensus formation
- A simple proof of the Cucker-Smale flocking dynamics and mean-field limit
- State transitions and the continuum limit for a 2D interacting, self-propelled particle system
- Flocking in noisy environments
- A conditional, collision-avoiding, model for swarming
- Avoiding Collisions in Flocks
- On collision-avoiding initial configurations to Cucker-Smale type flocking models
- On Krause's Multi-Agent Consensus Model With State-Dependent Connectivity
- A Cucker-Smale model with noise and delay
- Clustering and asymptotic behavior in opinion formation
- A conceptual model for milling formations in biological aggregates
- Modeling opinion dynamics: how the network enhances consensus
- Eulerian dynamics with a commutator forcing
- Inferring interaction rules from observations of evolutive systems I: The variational approach
- Active particles, Volume 1. Advances in theory, models, and applications
- Nonparametric inference of interaction laws in systems of agents from trajectory data
Cited In (11)
- Learning particle swarming models from data with Gaussian processes
- Learning interaction kernels in stochastic systems of interacting particles from multiple trajectories
- The discovery of dynamics via linear multistep methods and deep learning: error estimation
- Parameter calibration with stochastic gradient descent for interacting particle systems driven by neural networks
- Tools to Characterize the Correlated Nature of Collective Dynamics
- Learning theory for inferring interaction kernels in second-order interacting agent systems
- Data-driven entropic spatially inhomogeneous evolutionary games
- Data-Driven Model Selections of Second-Order Particle Dynamics via Integrating Gaussian Processes with Low-Dimensional Interacting Structures
- Learning Interaction Kernels in Mean-Field Equations of First-Order Systems of Interacting Particles
- Title not available (Why is that?)
- Nonparametric inference of interaction laws in systems of agents from trajectory data
This page was built for publication: Data-driven discovery of emergent behaviors in collective dynamics
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2127368)
