Nando de Freitas

Arnaud Doucet, Nando de Freitas & Neil Gordon (eds). Springer-Verlag. 2001. ISBN 0-387-95146-6.



Sequential Monte Carlo methods, also known as bootstrap filters, condensation, particle filters and survival of the fittest, have made it possible to numerically solve many complex, non-standard problems that were previously intractable. This book presented the first comprehensive and coherent 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.

When I began my work in this field in the late 90's, I realized that there were incomplete bits of the methodology in many fields. These fields weren't talking to each other. To make progress in all those fields, it was important to bring people together. So instead of writing the book, I chose to only write a few chapters and invite people in many other areas to contribute a chapter. It worked! The rest is history. Arnaud, Neil and I are pretty pleased with the results. The methodology is now more mature and coherent accross the various disciplines.

Having said this, there has been an incredible amount of progress in this area ... I'm still toying with the idea of a second book if I ever have the time ...

Last time I checked Citeseer, the book was the 5th most cited article in computer science in 2000. Current Citeseer rank