This paper details how the Bayesian-network structure of the posterior distribution of state-space models can be exploited to build improved parameterizations for system identification using variational inference. Three different parameterizations of the assumed state-path posterior distribution are proposed based on this representation: time-varying, steady-state, and convolution-smoother; each resulting in a different parameter estimation method. In contrast to existing methods for variational system identification, the proposed estimators can be implemented with unconstrained optimization methods. Furthermore, when applied to mini-batches in conjunction with stochastic optimization methods, the convolution-smoother formulation enables identification of large linear and nonlinear state-space systems from very large datasets. For linear systems, the method achieves the same performance as the inherently sequential prediction-error methods using and embarrassingly parallel algorithm that benefits from large speedups when computed in modern graphical processing units (GPUs). The ability of the proposed estimators to identify large models, work with large datasets split into mini-batches, and be work in parallel on GPUs make them well-suited for identifying deep models for applications in systems and control.
翻译:暂无翻译