Nonparametric Bayesian identification of jump systems with sparse dependencies
Name
Willsky_Nonparametric Bayesian.pdf
Size
609.17 KB
Format
Adobe PDF
Checksum (MD5)
af125fddb693547a19d64d9f09897f06
Author(s)
Fox, Emily Beth
Sudderth, Erik B.
Jordan, Michael I.
Willsky, Alan S.
Date Issued
July 2009
Journal
15th Symposium on System Identification, SYSID 2009
Publisher
International Federation of Automatic Control (IFAC)
Citation
Emily, Fox. “Nonparametric Bayesian Identification of Jump Systems with Sparse Dependencies.” Ed. Walter Eric. 15th Symposium on System Identification, SYSID 2009. 1591–1596. © 2009 IFAC
Version
Author's final manuscript
Abstract
Many nonlinear dynamical phenomena can be effectively modeled by a system that switches among a set of conditionally linear dynamical modes. We consider two such Markov jump linear systems: the switching linear dynamical system (SLDS) and the switching vector autoregressive (S-VAR) process. In this paper, we present a nonparametric Bayesian approach to identifying an unknown number of persistent, smooth dynamical modes by utilizing a hierarchical Dirichlet process prior. We additionally employ automatic relevance determination to infer a sparse set of dynamic dependencies. The utility and flexibility of our models are demonstrated on synthetic data and a set of honey bee dances.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike 3.0
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.3182/20090706-3-FR-2004.00264
