MAD-Bayes matching and alignment for labelled and unlabelled configurations

Publisher:
John Wiley & Sons
Publication Type:
Chapter
Citation:
Geometry Driven Statistics, 2015, pp. 377 - 390
Issue Date:
2015-11-27
Full metadata record
Files in This Item:
Filename Description Size
Green-MardiaFestschrift.pdfAccepted Manuscript version124.57 kB
Adobe PDF
Professor Kanti Mardia has made numerous original contributions to an area of unlabelled shape analysis inspired by matching and alignment problems arising in protein bioinformatics. In this chapter, a Bayesian model proposed by Mardia and Green in 2006, and others related to it, are revisited to investigate the potential for using modern optimisation algorithms to expedite calculations of properties of the posterior distribution, in place of the Monte Carlo computational methods originally proposed.
Please use this identifier to cite or link to this item: