Nonparametric inference of interaction laws in systems of agents from trajectory data
Authors:
Preprint
Abstract
Inferring the laws of interaction between particles and agents in complex dynamical systems from observational data is a fundamental challenge in a wide variety of disciplines. We propose a non-parametric statistical learning approach to estimate the governing laws of distance-based interactions, with no reference or assumption about their analytical form, from data consisting trajectories of interacting agents. We demonstrate the effectiveness of our learning approach both by providing theoretical guarantees, and by testing the approach on a variety of prototypical systems in various disciplines. These systems include homogeneous and heterogeneous agents systems, ranging from particle systems in fundamental physics to agent-based systems modeling opinion dynamics under the social influence, prey-predator dynamics, flocking and swarming, and phototaxis in cell dynamics.Keywords
All Related Versions
This publication has 55 references indexed in Scilit:
- Existence of Ground States of Nonlocal-Interaction EnergiesJournal of Statistical Physics, 2015
- Universal Algorithms for Learning Theory. Part II: Piecewise Polynomial FunctionsConstructive Approximation, 2007
- Transport Equations in Biology2007
- Approximation Methods for Supervised LearningFoundations of Computational Mathematics, 2005
- Modeling Language EvolutionFoundations of Computational Mathematics, 2004
- Self‐organization and selection in the emergence of vocabularyComplexity, 2002
- A Distribution-Free Theory of Nonparametric Regression2002
- Individual differences make a difference in the trajectories of simulated schools of fishEcological Modelling, 1996
- Self-organizing Dynamic Model of Fish SchoolingJournal of Theoretical Biology, 1994
- The use of splines with unknown end points in the identification of systemsJournal of Mathematical Analysis and Applications, 1971