ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Data-driven model reduction of agent-based systems using the Koopman generator | JoVE Visualize

Home
Research Domains
Information And Computing Sciences

Data-driven model reduction of agent-based systems using the Koopman generator

Jan-Hendrik Niemann^1,2, Stefan Klus^2,3, Christof Schütte^1,2

¹Modeling and Simulation of Complex Processes, Zuse Institute Berlin, Berlin, Germany.

Plos One|May 13, 2021

View abstract on PubMed

Summary

This study introduces Koopman operator theory to create simplified models of agent-based systems. This method efficiently captures complex social dynamics, even with many agents, by learning coarse-grained models from simulation data.

Area of Science:

Computational Social Science
Dynamical Systems Theory

Background:

Agent-based models (ABMs) simulate complex social systems.
Large-scale ABMs are computationally intensive.
Emergent behavior in ABMs arises from agent interactions.

Purpose of the Study:

To develop reduced-order models for large-scale agent-based systems.
To apply Koopman operator theory for model reduction.
To represent reduced dynamics using differential equations.

Main Methods:

Utilizing Koopman operator theory on ABM simulation data.
Learning coarse-grained models from empirical data.
Deriving aggregated state variables for collective behavior.

Main Results:

Successfully derived reduced models for agent-based systems.
Obtained reduced dynamics represented by differential equations.
Demonstrated good agreement with known analytical results for large agent populations.

Conclusions:

Koopman operator theory provides an effective method for ABM reduction.
The approach enables efficient analysis of complex social system dynamics.
This technique is valuable for studying collective behavior in large populations.

Artificial Intelligence

Fuzzy Computation

Data-driven Model Reduction Of Agent-based Systems Using The Koopman Generator