Bayesian function registration with random truncation
Open Access
- 7 July 2023
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 18 (7) , e0287734
Abstract
In this work, we develop a new set of Bayesian models to perform registration of real-valued functions. A Gaussian process prior is assigned to the parameter space of time warping functions, and a Markov chain Monte Carlo (MCMC) algorithm is utilized to explore the posterior distribution. While the proposed model can be defined on the infinite-dimensional function space in theory, dimension reduction is needed in practice because one cannot store an infinite-dimensional function on the computer. Existing Bayesian models often rely on some pre-specified, fixed truncation rule to achieve dimension reduction, either by fixing the grid size or the number of basis functions used to represent a functional object. In comparison, the new models in this paper randomize the truncation rule. Benefits of the new models include the ability to make inference on the smoothness of the functional parameters, a data-informative feature of the truncation rule, and the flexibility to control the amount of shape-alteration in the registration process. For instance, using both simulated and real data, we show that when the observed functions exhibit more local features, the posterior distribution on the warping functions automatically concentrates on a larger number of basis functions. Supporting materials including code and data to perform registration and reproduce some of the results presented herein are available online.Keywords
Funding Information
- National Science Foundation (CCF-1740761)
- National Science Foundation (CCF-1839252)
- National Science Foundation (DMS-2015226)
- Foundation for the National Institutes of Health (R37-CA214955)
This publication has 34 references indexed in Scilit:
- MCMC Methods for Functions: Modifying Old Algorithms to Make Them FasterStatistical Science, 2013
- A real-time system for biomechanical analysis of human movement and muscle functionMedical & Biological Engineering & Computing, 2013
- Statistical Modeling of Curves Using Shapes and Related FeaturesJournal of the American Statistical Association, 2012
- Methods to temporally align gait cycle dataJournal of Biomechanics, 2011
- A Multiresolution Approach to Time Warping Achieved by a Bayesian Prior–Posterior Transfer Fitting StrategyJournal of the Royal Statistical Society Series B: Statistical Methodology, 2010
- Bayesian Hierarchical Curve RegistrationJournal of the American Statistical Association, 2008
- Curve registration by local regressionCanadian Journal of Statistics, 2000
- Functional Components of Variation in HandwritingJournal of the American Statistical Association, 2000
- Curve RegistrationJournal of the Royal Statistical Society Series B: Statistical Methodology, 1998
- A Functional Data Analysis of the Pinch Force of Human FingersJournal of the Royal Statistical Society Series C: Applied Statistics, 1995