Amortized backward variational inference in nonlinear state-space models

Abstract

We consider the problem of state estimation in general state-space models using variational inference. For a generic variational family defined using the same backward decomposition as the actual joint smoothing distribution, we establish for the first time that, under mixing assumptions, the variational approximation of expectations of additive state functionals induces an error which grows at most linearly in the number of observations. This guarantee is consistent with the known upper bounds for the approximation of smoothing distributions using standard Monte Carlo methods. Moreover, we propose an amortized inference framework where a neural network shared over all times steps outputs the parameters of the variational kernels. We also study empirically parametrizations which allow analytical marginalization of the variational distributions, and therefore lead to efficient smoothing algorithms. Significant improvements are made over state-of-the art variational solutions, especially when the generative model depends on a strongly nonlinear and noninjective mixing function.

References in corpus (8)

• Auto-Encoding Variational Bayes
• Composing graphical models with neural networks for structured representations and fast inference
• Sequential Monte Carlo smoothing for general state space hidden Markov models
• Variational Autoencoders and Nonlinear ICA: A Unifying Framework
• Weakly-Supervised Disentanglement Without Compromises
• Disentangling Identifiable Features from Noisy Data with Structured Nonlinear ICA
• Fast and numerically stable particle-based online additive smoothing: the AdaSmooth algorithm
• Identifiability and consistent estimation of nonparametric translation hidden Markov models with general state space

Cited by in corpus (1)

• Variational excess risk bound for general state space models