Double-Chain Unscented Expectation Propagation for Inference in Stochastic Dynamical Models of Biological Processes

Hao Wu, Stefan Bernhard
2013 International Work-Conference on Bioinformatics and Biomedical Engineering  
In many fields of biology and medicine we are confronted with the task of analyzing and estimating a complex process based on partial observations. Many of these problems can be described by dynamical system models with unknown states and parameters. The Bayesian theory provides a statistical inference framework for such systems. But there is a fundamental and not satisfactorily solved problem in Bayesian inference of biological processes, that is how to approximate the posterior distributions
more » ... ast and accurately in real world applications. Here we develop a new algorithm within the expectation propagation framework to solve this problem. The usefulness of the method will be demonstrated by applying it to two significantly different biomedically important problems: single-molecule fluorescence experiments and cardiovascular system measurements.
dblp:conf/iwbbio/WuB13 fatcat:m3x43kqf6jadbl72lvanl25dh4