Sequential Monte Carlo Methods in Practice (Information Science and Statistics)

Monte Carlo methods are revolutionising the on-line analysis of data in fields as diverse as financial modelling, target tracking and computer vision. These methods, appearing under the names of bootstrap filters, condensation, optimal Monte Carlo filters, particle filters and survial of the fittest, have made it possible to solve numerically many complex, non-standarard problems that were previously intractable. This book presents the first comprehensive treatment of these techniques, including convergence results and applications to tracking, guidance, automated target recognition, aircraft navigation, robot navigation, econometrics, financial modelling, neural networks, optimal control, optimal filtering, communications, reinforcement learning, signal enhancement, model averaging and selection, computer vision, semiconductor design, population biology, dynamic Bayesian networks, and time series analysis. This will be of great value to students, researchers and practicioners, who have some basic knowledge of probability. Arnaud Doucet received the Ph. D. degree from the University of Paris- XI Orsay in 1997. From 1998 to 2000, he conducted research at the Signal Processing Group of Cambridge University, UK. He is currently an assistant professor at the Department of Electrical Engineering of Melbourne University, Australia. His research interests include Bayesian statistics, dynamic models and Monte Carlo methods. Nando de Freitas obtained a Ph. D. degree in information engineering from Cambridge University in 1999. He is presently a research associate with the artificial intelligence group of the University of California at Berkeley. His main research interests are in Bayesian statistics and the application of on-line and batch Monte Carlo methods to machine learning "Monte Carlo methods are revolutionizing the on-line analysis of data in fields as diverse as financial modeling, target tracking and computer vision. These methods, appearing under the names of bootstrap filters, condensation, optimal Monte Carlo filters, particle filters and survival of the fittest, have made it possible to solve numerically many complex, non-standard problems that were previously intractable. This book presents the first comprehensive treatment of these techniques, including convergence results and applications to tracking, guidance, automated target recognition, aircraft navigation, robot navigation, econometrics, financial modeling, neural networks, optimal control, optimal filtering, communications, reinforcement learning, signal enhancement, model averaging and selection, computer vision, semiconductor design, population biology, dynamic Bayesian networks, and time series analysis. This will be of great value to students, researchers and practitioners, who have some basic knowledge of probability. Arnaud Doucet received the Ph. D. degree from the University of Paris-XI Orsay in 1997. From 1998 to 2000, he conducted research at the Signal Processing Group of Cambridge University, UK. He is currently an assistant professor at the Department of Electrical Engineering of Melbourne University, Australia. His research interests include Bayesian statistics, dynamic models and Monte Carlo methods. Nando de Freitas obtained a Ph. D. degree in information engineering from Cambridge University in 1999. He is presently a research associate with the artificial intelligence group of the University of California at Berkeley. His main research interests are in Bayesian statistics and the application of on-line and batch Monte Carlo methods to machine learning. Neil Gordon obtained a Ph. D. in Statistics from Imperial College, University of London in 1993. He is with the Pattern and Information Processing group at the Defence Evaluation and Research Agency in the United Kingdom. His research interests are in time series, statistical data analysis, and pattern recognition with a particular emphasis on target tracking and missile guidance."-- Provided by publisher Front Matter....Pages i-xxvii Front Matter....Pages 1-1 An Introduction to Sequential Monte Carlo Methods....Pages 3-14 Front Matter....Pages 15-15 Particle Filters — A Theoretical Perspective....Pages 17-41 Interacting Particle Filtering With Discrete Observations....Pages 43-75 Front Matter....Pages 77-77 Sequential Monte Carlo Methods for Optimal Filtering....Pages 79-95 Deterministic and Stochastic Particle Filters in State-Space Models....Pages 97-116 RESAMPLE-MOVE Filtering with Cross-Model Jumps....Pages 117-138 Improvement Strategies for Monte Carlo Particle Filters....Pages 139-158 Approximating and Maximising the Likelihood for a General State-Space Model....Pages 159-175 Monte Carlo Smoothing and Self-Organising State-Space Model....Pages 177-195 Combined Parameter and State Estimation in Simulation-Based Filtering....Pages 197-223 A Theoretical Framework for Sequential Importance Sampling with Resampling....Pages 225-246 Improving Regularised Particle Filters....Pages 247-271 Auxiliary Variable Based Particle Filters....Pages 273-293 Improved Particle Filters and Smoothing....Pages 295-317 Front Matter....Pages 319-319 Posterior Cramér-Rao Bounds for Sequential Estimation....Pages 321-338 Statistical Models of Visual Shape and Motion....Pages 339-357 Sequential Monte Carlo Methods for Neural Networks....Pages 359-379 Sequential Estimation of Signals under Model Uncertainty....Pages 381-400 Particle Filters for Mobile Robot Localization....Pages 401-428 Self-Organizing Time Series Model....Pages 429-444 Front Matter....Pages 319-319 Sampling in Factored Dynamic Systems....Pages 445-464 In-Situ Ellipsometry Solutions Using Sequential Monte Carlo....Pages 465-477 Manoeuvring Target Tracking Using a Multiple-Model Bootstrap Filter....Pages 479-497 Rao-Blackwellised Particle Filtering for Dynamic Bayesian Networks....Pages 499-515 Particles and Mixtures for Tracking and Guidance....Pages 517-532 Monte Carlo Techniques for Automated Target Recognition....Pages 533-552 Back Matter....Pages 553-582 "Monte Carlo methods are revolutionizing the on-line analysis of data in fields as diverse as financial modelling, target tracking, and computer vision. These methods, appearing under the names of bootstrap filters, condensation, optimal Monte Carlo filters, particle filters, and survival of the fittest, have made it possible to solve numerically many complex, nonstandard problems that were previously intractable. This book presents the first comprehensive and coherent treatment of these techniques, including convergence results and applictions to tracking, guidance, automated target recognition, aircraft navigation, robot navigation, economometrics, financial modelling, neural networks, optimal control, optimal filtering, communicaitons, reinforcement learning, signal enhancement, model averaging and selection, computer vision, semiconductor design, population biology, dynamic Bayesian networks, and time series analysis ... ." (source : 4ème de couverture)

Monte Carlo methods are revolutionizing the on-line analysis of data in many fileds. They have made it possible to solve numerically many complex, non-standard problems that were previously intractable. This book presents the first comprehensive treatment of these techniques.