TY - JOUR
T1 - Variational Bayes identification of acyclic dynamic networks
AU - Risuleo, R.S.
AU - Bottegal, G.
AU - Hjalmarsson, H.
PY - 2017/7/1
Y1 - 2017/7/1
N2 - We study the problem of identifying dynamic networks that do not present loops. We model the impulse responses of the modules in the network as zero-mean independent Gaussian processes. The covariance matrices of the processes can be used to encode prior information, such as stability and smoothness, about the impulse responses of the modules. To estimate the modules, we approximate the joint posterior distribution of the impulse responses using a variational Bayes approach. In particular, using a mean-field approximation, we assume a factorization of the posterior where each factor corresponds to a single module. We estimate the kernel hyperparameters and the measurement noise variances by combining variational Bayes with the expectation-maximization method. We evaluate the performance of the identification procedure in a simulation experiment, where we compare to other kernel-based approaches.
AB - We study the problem of identifying dynamic networks that do not present loops. We model the impulse responses of the modules in the network as zero-mean independent Gaussian processes. The covariance matrices of the processes can be used to encode prior information, such as stability and smoothness, about the impulse responses of the modules. To estimate the modules, we approximate the joint posterior distribution of the impulse responses using a variational Bayes approach. In particular, using a mean-field approximation, we assume a factorization of the posterior where each factor corresponds to a single module. We estimate the kernel hyperparameters and the measurement noise variances by combining variational Bayes with the expectation-maximization method. We evaluate the performance of the identification procedure in a simulation experiment, where we compare to other kernel-based approaches.
KW - Bayesian methods
KW - Nonparametric methods
KW - System identification
UR - http://www.scopus.com/inward/record.url?scp=85031801139&partnerID=8YFLogxK
U2 - 10.1016/j.ifacol.2017.08.1318
DO - 10.1016/j.ifacol.2017.08.1318
M3 - Conference article
AN - SCOPUS:85031801139
SN - 2405-8963
VL - 50
SP - 10556
EP - 10561
JO - IFAC-PapersOnLine
JF - IFAC-PapersOnLine
IS - 1
ER -