Data-driven Control of Agent-based Models: an Equation/Variable-free Machine Learning Approach

Abstract

We present an Equation/Variable free machine learning (EVFML) framework for the control of the collective dynamics of complex/multiscale systems modelled via microscopic/agent-based simulators. The approach obviates the need for construction of surrogate, reduced-order models.~The proposed implementation consists of three steps: (A) from high-dimensional agent-based simulations, machine learning (in particular, non-linear manifold learning (Diffusion Maps (DMs)) helps identify a set of coarse-grained variables that parametrize the low-dimensional manifold on which the emergent/collective dynamics evolve. The out-of-sample extension and pre-image problems, i.e. the construction of non-linear mappings from the high-dimensional input space to the low-dimensional manifold and back, are solved by coupling DMs with the Nystrom extension and Geometric Harmonics, respectively; (B) having identified the manifold and its coordinates, we exploit the Equation-free approach to perform numerical bifurcation analysis of the emergent dynamics; then (C) based on the previous steps, we design data-driven embedded wash-out controllers that drive the agent-based simulators to their intrinsic, imprecisely known, emergent open-loop unstable steady-states, thus demonstrating that the scheme is robust against numerical approximation errors and modelling uncertainty.~The efficiency of the framework is illustrated by controlling emergent unstable (i) traveling waves of a deterministic agent-based model of traffic dynamics, and (ii) equilibria of a stochastic financial market agent model with mimesis.

References in corpus (25)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
• DeepXDE: A deep learning library for solving differential equations
• A Data-Driven Approximation of the Koopman Operator: Extending Dynamic Mode Decomposition
• Solving high-dimensional partial differential equations using deep learning
• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Deep learning to represent sub-grid processes in climate models
• Control Principles of Complex Networks
• Koopman invariant subspaces and finite linear representations of nonlinear dynamical systems for control
• Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration
• Local Extreme Learning Machines and Domain Decomposition for Solving Linear and Nonlinear Partial Differential Equations
• Deep learning observables in computational fluid dynamics
• Deep least-squares methods: an unsupervised learning-based numerical method for solving elliptic PDEs
• Coarse-scale PDEs from fine-scale observations via machine learning
• Systematic experimental exploration of bifurcations with non-invasive control
• On Computing the Hyperparameter of Extreme Learning Machines: Algorithm and Application to Computational PDEs, and Comparison with Classical and High-Order Finite Elements
• Reduced models in chemical kinetics via nonlinear data-mining
• Analysis of the accuracy and convergence of equation-free projection to a slow manifold
• Constructing coarse-scale bifurcation diagrams from spatio-temporal observations of microscopic simulations: A parsimonious machine learning approach
• Manifold learning for parameter reduction
• Implicit Methods for Equation-Free Analysis: Convergence Results and Analysis of Emergent Waves in Microscopic Traffic Models
• Numerical Continuation in Nonlinear Experiments using Local Gaussian Process Regression
• Coarse Bifurcation Diagrams via Microscopic Simulators: A State-Feedback Control-Based Approach
• On the Koopman operator of algorithms
• Time Series Forecasting Using Manifold Learning

Cited by in corpus (9)

• Continuification-based control of large multiagent systems in a ring
• Data-driven modelling of brain activity using neural networks, Diffusion Maps, and the Koopman operator
• Nonlinear Discrete-Time Observers with Physics-Informed Neural Networks
• Slow Invariant Manifolds of Singularly Perturbed Systems via Physics-Informed Machine Learning
• GoRINNs: Godunov-Riemann Informed Neural Networks for Learning Hyperbolic Conservation Laws
• Cell reprogramming design by transfer learning of functional transcriptional networks
• A physics-informed neural network method for the approximation of slow invariant manifolds for the general class of stiff systems of ODEs
• Learning the Latent dynamics of Fluid flows from High-Fidelity Numerical Simulations using Parsimonious Diffusion Maps
• A CRISP-DM-based Methodology for Assessing Agent-based Simulation Models using Process Mining