Atlas home || Conferences | Abstracts | about Atlas

International Conference on Statistics, Combinatorics and Related Areas and the Eighth International Conference of Forum for Interdisciplinary Mathematics
December 19-21, 2001
School of Mathematics and Applied Statistics, University of Wollongong
Wollongong, NSW, Australia

Organizers
Satya N. Mishra (University of South Alabama), Chandra M. Gulati (University of Wollongong)

View Abstracts
Conference Homepage

Multiplicative correction algorithms for positive MPL in large linear inverse problems
by
Jun Ma
Macquarie University

In many linear inverse problems the unknown function which needs to be reconstructed is subject to the positive constraint. How to develop an easy-to-implement algorithm for the maximum penalized likelihood (MPL) estimates for positive inverse problems is an interesting and challenging question. Given that observations and point-spread-function (PSF) values are all non-negative, we propose a simple iterative method. This method is based on the multiplicative correction scheme, but it can be rearranged to appear as a gradient algorithm. The new algorithm is developed in the context where observations are subject to general noise distributions, i.e. distributions in the exponential family. The convergence properties of the algorithm are studied in this paper.

Date received: November 21, 2001

Copyright © 2001 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # caim-03.