A Machine Learning Framework for Solving High-Dimensional Mean Field Game and Mean Field Control Problems

Abstract

Mean field games (MFG) and mean field control (MFC) are critical classes of multi-agent models for efficient analysis of massive populations of interacting agents. Their areas of application span topics in economics, finance, game theory, industrial engineering, crowd motion, and more. In this paper, we provide a flexible machine learning framework for the numerical solution of potential MFG and MFC models. State-of-the-art numerical methods for solving such problems utilize spatial discretization that leads to a curse-of-dimensionality. We approximately solve high-dimensional problems by combining Lagrangian and Eulerian viewpoints and leveraging recent advances from machine learning. More precisely, we work with a Lagrangian formulation of the problem and enforce the underlying Hamilton-Jacobi-Bellman (HJB) equation that is derived from the Eulerian formulation. Finally, a tailored neural network parameterization of the MFG/MFC solution helps us avoid any spatial discretization. Our numerical results include the approximate solution of 100-dimensional instances of optimal transport and crowd motion problems on a standard work station and a validation using an Eulerian solver in two dimensions. These results open the door to much-anticipated applications of MFG and MFC models that were beyond reach with existing numerical methods.

21 pages, 13 figures, 2 table

References in corpus (13)

• DGM: A deep learning algorithm for solving partial differential equations
• Solving high-dimensional partial differential equations using deep learning
• Variational Inference with Normalizing Flows
• Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
• Visualizing the Loss Landscape of Neural Nets
• Deep Learning Approximation for Stochastic Control Problems
• Maximum Principle Based Algorithms for Deep Learning
• A Lagrangian Gauss-Newton-Krylov Solver for Mass- and Intensity-Preserving Diffeomorphic Image Registration
• A mean-field game approach to price formation in electricity markets
• Size estimates for fat inclusions in an isotropic Reissner-Mindlin plate
• Fluid Flow Mass Transport for Generative Networks
• Potential Flow Generator with $L_2$ Optimal Transport Regularity for Generative Models
• Constrained dynamical optimal transport and its Lagrangian formulation

Cited by in corpus (44)

• Algorithms for Solving High Dimensional PDEs: From Nonlinear Monte Carlo to Machine Learning
• How to train your neural ODE: the world of Jacobian and kinetic regularization
• An overview on deep learning-based approximation methods for partial differential equations
• A Discontinuity Capturing Shallow Neural Network for Elliptic Interface Problems
• Learning Differential Equations that are Easy to Solve
• In-Context Operator Learning with Data Prompts for Differential Equation Problems
• A Neural Network Approach for High-Dimensional Optimal Control Applied to Multi-Agent Path Finding
• Model-Free Mean-Field Reinforcement Learning: Mean-Field MDP and Mean-Field Q-Learning
• A Machine Learning Framework for Computing the Most Probable Paths of Stochastic Dynamical Systems
• Derivative-Informed Projected Neural Networks for High-Dimensional Parametric Maps Governed by PDEs
• Dynamic tensor approximation of high-dimensional nonlinear PDEs
• Numerical methods for Mean field Games based on Gaussian Processes and Fourier Features
• A Neural Network Approach Applied to Multi-Agent Optimal Control
• Bridging Mean-Field Games and Normalizing Flows with Trajectory Regularization
• Sparse Gaussian processes for solving nonlinear PDEs
• Neural networks-based algorithms for stochastic control and PDEs in finance
• Convergence of flow-based generative models via proximal gradient descent in Wasserstein space
• Memory-Limited Partially Observable Stochastic Control and its Mean-Field Control Approach
• High order computation of optimal transport, mean field planning, and mean field games
• Computational Mean-field Games on Manifolds
• Random Features for High-Dimensional Nonlocal Mean-Field Games
• MALI: A memory efficient and reverse accurate integrator for Neural ODEs
• Pontryagin's Minimum Principle and Forward-Backward Sweep Method for the System of HJB-FP Equations in Memory-Limited Partially Observable Stochastic Control
• A Simple Multiscale Method for Mean Field Games
• Integrating Dynamical Systems Modeling with Spatiotemporal scRNA-seq Data Analysis
• Discovering Efficient Periodic Behaviours in Mechanical Systems via Neural Approximators
• On the Representation of Solutions to Elliptic PDEs in Barron Spaces
• Mean Field LQG Social Optimization: A Reinforcement Learning Approach
• Optimal Dirichlet Boundary Control by Fourier Neural Operators Applied to Nonlinear Optics
• Mean-Field Control Approach to Decentralized Stochastic Control with Finite-Dimensional Memories
• Alternating the Population and Control Neural Networks to Solve High-Dimensional Stochastic Mean-Field Games
• Learning High Dimensional Wasserstein Geodesics
• Theory for Optimal Estimation and Control under Resource Limitations and Its Applications to Biological Information Processing and Decision-Making
• A Bilevel Optimization Method for Inverse Mean-Field Games
• Signatured Deep Fictitious Play for Mean Field Games with Common Noise
• Resource Limitations induce Phase Transitions in Biological Information Processing
• Go with the Flow: Adaptive Control for Neural ODEs
• Deep Learning for Mean Field Games and Mean Field Control with Applications to Finance
• Machine Learning for Discovering Effective Interaction Kernels between Celestial Bodies from Ephemerides
• Learning ODEs via Diffeomorphisms for Fast and Robust Integration
• A Blob Method for Mean Field Control With Terminal Constraints
• A Fast Proximal Gradient Method and Convergence Analysis for Dynamic Mean Field Planning
• Approximating the Optimal Transport Plan via Particle-Evolving Method
• Deep Semi-Martingale Optimal Transport