Message Passing-Based Inference for Time-Varying Autoregressive Models |
| |
Authors: | Albert Podusenko Wouter M Kouw Bert de Vries |
| |
Institution: | 1.Department of Electrical Engineering, Eindhoven University of Technology, 5612 AZ Eindhoven, The Netherlands; (W.M.K.); (B.d.V.);2.GN Hearing, JF Kennedylaan 2, 5612 AB Eindhoven, The Netherlands |
| |
Abstract: | Time-varying autoregressive (TVAR) models are widely used for modeling of non-stationary signals. Unfortunately, online joint adaptation of both states and parameters in these models remains a challenge. In this paper, we represent the TVAR model by a factor graph and solve the inference problem by automated message passing-based inference for states and parameters. We derive structured variational update rules for a composite “AR node” with probabilistic observations that can be used as a plug-in module in hierarchical models, for example, to model the time-varying behavior of the hyper-parameters of a time-varying AR model. Our method includes tracking of variational free energy (FE) as a Bayesian measure of TVAR model performance. The proposed methods are verified on a synthetic data set and validated on real-world data from temperature modeling and speech enhancement tasks. |
| |
Keywords: | Bayesian inference free energy factor graph hybrid message passing model selection non-stationary systems probabilistic graphical models |
|
|