On the concentration properties of interacting particle processes

DOI10.1561/2200000026zbMATH Open1280.60057arXiv1107.1948OpenAlexW2119909095MaRDI QIDQ2903299FDOQ2903299

Authors: Pierre Del Moral, Peng Hu, Liming Wu 

Publication date: 8 August 2012

Published in: Foundations and Trends in Machine Learning (Search for Journal in Brave)

Abstract: These lecture notes present some new concentration inequalities for Feynman-Kac particle processes. We analyze different types of stochastic particle models, including particle profile occupation measures, genealogical tree based evolution models, particle free energies, as well as backward Markov chain particle models. We illustrate these results with a series of topics related to computational physics and biology, stochastic optimization, signal processing and bayesian statistics, and many other probabilistic machine learning algorithms. Special emphasis is given to the stochastic modeling and the quantitative performance analysis of a series of advanced Monte Carlo methods, including particle filters, genetic type island models, Markov bridge models, interacting particle Markov chain Monte Carlo methodologies.

Full work available at URL: https://arxiv.org/abs/1107.1948

Recommendations

• Convergence of \(U\)-statistics for interacting particle systems
• A limit theorem for a class of interacting particle systems
• Concentration of measure for Brownian particle systems interacting through their ranks
• Concentration inequalities for mean field particle models
• scientific article; zbMATH DE number 1236226
• scientific article; zbMATH DE number 1405914
• scientific article; zbMATH DE number 6492985
• scientific article; zbMATH DE number 1788308

zbMATH Keywords

Bayesian statisticsstochastic optimizationconcentration inequalitiesadvanced Monte Carlo methodscomputational physics and biologyFeynman-Kac particle processes

Mathematics Subject Classification ID

Asymptotic properties of nonparametric inference (62G20) Monte Carlo methods (65C05) Inequalities; stochastic orderings (60E15) Interacting random processes; statistical mechanics type models; percolation theory (60K35) Stochastic particle methods (65C35) Interacting particle systems in time-dependent statistical mechanics (82C22) Schrödinger and Feynman-Kac semigroups (47D08)

Cited In (13)

• Variations on Hammersley’s interacting particle process
• Sensitivity analysis for rare events based on Rényi divergence
• Annealed Feynman-Kac models
• Particle methods: an introduction with applications
• Robust filtering: correlated noise and multidimensional observation
• Limit theorems for cloning algorithms
• Concentration inequalities for mean field particle models
• Feynman-Kac particle integration with geometric interacting jumps
• Geometric Convergence and Concentration Inequalities for the Feynman–Kac Genetic Algorithm
• Nonasymptotic analysis of adaptive and annealed Feynman-Kac particle models
• On the mean shape of particle processes
• Convergence of the one-dimensional contact process with two types of particles and priority
• On the foundations and the applications of evolutionary computing

This page was built for publication: On the concentration properties of interacting particle processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2903299)