A class of dimension-free metrics for the convergence of empirical measures

DOI10.1016/J.SPA.2023.07.009MaRDI QIDQ6072907zbMATH OpenOpenAlexFDO

Authors Jiequn Han, Ruimeng Hu, Jihao Long

Publication date 15 September 2023

Published in Stochastic Processes and their Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2104.12036

zbMATH Keywords

curse of dimensionalitymean-field gamesempirical measureintegral probability metricsMckean-Vlasov stochastic differential equation

Mathematics Subject Classification ID

Convergence of probability measures (60B10) Inequalities; stochastic orderings (60E15) Interacting random processes; statistical mechanics type models; percolation theory (60K35) Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Mean field games (aspects of game theory) (91A16)

Abstract: This paper concerns the convergence of empirical measures in high dimensions. We propose a new class of metrics and show that under such metrics, the convergence is free of the curse of dimensionality (CoD). Such a feature is critical for high-dimensional analysis and stands in contrast to classical metrics ({it e.g.}, the Wasserstein distance). The proposed metrics originate from the maximum mean discrepancy, which we generalize by proposing specific criteria for selecting test function spaces to guarantee the property of being free of CoD. Therefore, we call this class of metrics the generalized maximum mean discrepancy (GMMD). Examples of the selected test function spaces include the reproducing kernel Hilbert space, Barron space, and flow-induced function spaces. Three applications of the proposed metrics are presented: 1. The convergence of empirical measure in the case of random variables; 2. The convergence of $n$-particle system to the solution to McKean-Vlasov stochastic differential equation; 3. The construction of an $varepsilon$-Nash equilibrium for a homogeneous $n$-player game by its mean-field limit. As a byproduct, we prove that, given a distribution close to the target distribution measured by GMMD and a certain representation of the target distribution, we can generate a distribution close to the target one in terms of the Wasserstein distance and relative entropy. Overall, we show that the proposed class of metrics is a powerful tool to analyze the convergence of empirical measures in high dimensions without CoD.

Recommendations

• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• Convergence and concentration of empirical measures under Wasserstein distance in unbounded functional spaces
• The CLT in high dimensions: quantitative bounds via martingale embedding
• A new framework for distance and kernel-based metrics in high dimensions
• Empirical measures: regularity is a counter-curse to dimensionality

Cites work

• scientific article; zbMATH DE number 1022519 (Why is no real title available?)
• scientific article; zbMATH DE number 1405266 (Why is no real title available?)
• scientific article; zbMATH DE number 3202900 (Why is no real title available?)
• scientific article; zbMATH DE number 3252891 (Why is no real title available?)
• 10.1162/153244303321897690
• A CLASS OF MARKOV PROCESSES ASSOCIATED WITH NONLINEAR PARABOLIC EQUATIONS
• A mean field view of the landscape of two-layer neural networks
• A priori estimates of the population risk for two-layer neural networks
• Advanced Lectures on Machine Learning
• Backward stochastic differential equations. From linear to fully nonlinear theory
• Combinatorics of random processes and sections of convex bodies
• Convergence of deep fictitious play for stochastic differential games
• Deep fictitious play for stochastic differential games
• From the master equation to mean field game limit theory: large deviations and concentration of measure
• Functional analysis. Introduction to further topics in analysis
• Generalization error of GAN from the discriminator's perspective
• Hanson-Wright inequality and sub-Gaussian concentration
• High-dimensional probability. An introduction with applications in data science
• Integral Probability Metrics and Their Generating Classes of Functions
• Kernel methods in machine learning
• Kolmogorov width decay and poor approximators in machine learning: shallow neural networks, random feature models and neural tangent kernels
• Large population stochastic dynamic games: closed-loop McKean-Vlasov systems and the Nash certainty equivalence principle
• Large-Population Cost-Coupled LQG Problems With Nonuniform Agents: Individual-Mass Behavior and Decentralized $\varepsilon$-Nash Equilibria
• Liquidity, risk measures, and concentration of measure
• Mean field analysis of neural networks: a law of large numbers
• Mean field games
• Mean field games. I: The stationary case
• Mean field games. II: Finite horizon and optimal control
• On a strong form of propagation of chaos for McKean-Vlasov equations
• On the moments of the modulus of continuity of Itô processes
• On the rate of convergence in Wasserstein distance of the empirical measure
• Probabilistic analysis of mean-field games
• Probabilistic theory of mean field games with applications I. Mean field FBSDEs, control, and games
• Probabilistic theory of mean field games with applications II. Mean field games with common noise and master equations
• Probability Metrics
• Propagation of chaos: a review of models, methods and applications. I: Models and methods
• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• Simulation of McKean-Vlasov SDEs with super-linear growth
• Some results on Tchebycheffian spline functions and stochastic processes
• Some stochastic particle methods for nonlinear parabolic PDEs
• The Barron space and the flow-induced function spaces for neural network models
• The Speed of Mean Glivenko-Cantelli Convergence
• Theory of Reproducing Kernels
• Transport Inequalities. A Survey
• Transportation cost-information inequalities and applications to random dynamical systems and diffusions.
• Understanding machine learning. From theory to algorithms
• Universal approximation bounds for superpositions of a sigmoidal function

Cited in

(2)

• Learning High-Dimensional McKean–Vlasov Forward-Backward Stochastic Differential Equations with General Distribution Dependence
• Recent developments in machine learning methods for stochastic control and games

This page was built for publication: A class of dimension-free metrics for the convergence of empirical measures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6072907)