PIONM: A Generalized Approach to Solving Density-Constrained Mean-Field Games Equilibrium under Modified Boundary Conditions

2025-04-04

AI Summary

This work addresses the computational bottleneck in solving high-dimensional density-constrained mean-field games (MFGs), where conventional methods require retraining whenever boundary conditions—such as initial distributions, terminal value functions, or obstacle configurations—change. We propose PIONM, a physics-informed neural operator-driven framework for universal MFG equilibrium computation. PIONM innovatively integrates physics-informed neural operators with discrete-time normalized flows, encoding boundary conditions as learnable feature embeddings to enable end-to-end solution of the coupled PDE system. Crucially, it achieves zero-shot generalization to unseen initial/terminal distributions and arbitrary obstacle layouts after a single training phase. Extensive experiments demonstrate significantly improved satisfaction of density constraints under diverse boundary perturbations, along with superior computational efficiency and scalability compared to state-of-the-art approaches.

Technology Category

• Multiagent Systems: Adversarial Agents
• Search and Optimization: Mixed Discrete/Continuous Search
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Neural network-based methods are effective for solving equilibria in Mean-Field Games (MFGs), particularly in high-dimensional settings. However, solving the coupled partial differential equations (PDEs) in MFGs limits their applicability since solving coupled PDEs is computationally expensive. Additionally, modifying boundary conditions, such as the initial state distribution or terminal value function, necessitates extensive retraining, reducing scalability. To address these challenges, we propose a generalized framework, PIONM (Physics-Informed Neural Operator NF-MKV Net), which leverages physics-informed neural operators to solve MFGs equations. PIONM utilizes neural operators to compute MFGs equilibria for arbitrary boundary conditions. The method encodes boundary conditions as input features and trains the model to align them with density evolution, modeled using discrete-time normalizing flows. Once trained, the algorithm efficiently computes the density distribution at any time step for modified boundary condition, ensuring efficient adaptation to different boundary conditions in MFGs equilibria. Unlike traditional MFGs methods constrained by fixed coefficients, PIONM efficiently computes equilibria under varying boundary conditions, including obstacles, diffusion coefficients, initial densities, and terminal functions. PIONM can adapt to modified conditions while preserving density distribution constraints, demonstrating superior scalability and generalization capabilities compared to existing methods.

Problem

• Solves density-constrained Mean-Field Games equilibria efficiently
• Handles modified boundary conditions without extensive retraining
• Generalizes to varying coefficients and obstacles in MFGs

Innovation

• Physics-informed neural operators solve MFGs equations
• Neural operators handle arbitrary boundary conditions
• Discrete-time normalizing flows model density evolution