Nonlinear state-space model identification is inherently challenging due to the need for joint latent-state inference and parameter learning. Variational inference offers a tractable framework, where structured Gaussian posteriors enable scalable inference over latent trajectories. However, existing parameterizations of the posterior mean struggle to capture complex nonlinear dynamics, while more expressive deep parameterizations tend to introduce instability in joint optimization. To address these issues, this article proposes a deep variational identification method for nonlinear state-space models based on a structured Gaussian posterior. Specifically, the posterior mean is parameterized using a non-causal dilated residual convolutional network, while a Markov structure with block-tridiagonal precision is preserved to ensure linear-time inference. Furthermore, an alternating optimization scheme is developed to separate variational smoothing from model identification. The variational parameters are updated by maximizing a differentiable approximation to the evidence lower bound, whereas the model and noise parameters are updated via analytic identification steps using samples from the variational posterior. Experiments on a nonlinear discrete-time system and a stochastic Duffing oscillator demonstrate that the proposed approach achieves stable optimization and reliable estimation of the system dynamics and noise statistics.
- Open Access
- Article
Deep Variational System Identification for Nonlinear State-Space Models via Alternating Optimization
- Kaiqi Fang 1,
- Guijun Ma 1,
- Yasen Wang 2,
- Zuogong Yue 1,
- Xiaoyou Duan 1,
- Maolin Wang 1,*
Author Information
Received: 31 Mar 2026 | Revised: 04 Jun 2026 | Accepted: 03 Aug 2026 | Published: 17 Aug 2026
Abstract
Keywords
nonlinear state-space models | system identification | Gaussian variational inference | structured Gaussian posterior | alternating optimization
References
- 1.
Ljung, L. System identification. In Signal Analysis and Prediction; Birkhauser: Boston, MA, USA, 1998; pp. 163–173.
- 2.
Yu, C.; Ljung, L.; Verhaegen, M. Identification of structured state-space models. Automatica 2018, 90, 54–61. https://doi.org/10.1016/j.automatica.2017.12.023.
- 3.
Shen, T.; Dong, Y.; He, D.; et al. Online identification of time-varying dynamical systems for industrial robots based on sparse Bayesian learning. Sci. China Technol. Sci. 2022, 65, 386–395. https://doi.org/10.1007/s11431-021-1916-0.
- 4.
Chen, S.; Werling, K.; Wu, A.; et al. Real-time model predictive control and system identification using differentiable simulation. IEEE Robot. Autom. Lett. 2023, 8, 312–319. https://doi.org/10.1109/LRA.2022.3213009.
- 5.
Marmarelis, V.Z. Nonlinear Dynamic Modeling of Physiological Systems; John Wiley & Sons: Hoboken, NJ, USA, 2004.
- 6.
Katayama, T. Subspace Methods for System Identification; Springer: London, UK, 2005.
- 7.
Verhaegen, M.; Verdult, V. Filtering and System Identification: A Least Squares Approach; Cambridge University Press: Cambridge, UK, 2007.
- 8.
Wang, Y.; Fang, K.; Ma, G.; et al. Learning linear state-space models with sparse system matrices. In Proceedings of the International Conference on Learning Representations, Rio de Janeiro, Brazil, 23–27 April 2026.
- 9.
Ljung, L. Perspectives on system identification. Annu. Rev. Control 2010, 34, 1–12. https://doi.org/10.1016/j.arcontrol.2009.12.001.
- 10.
Ninness, B. Some system identification challenges and approaches. IFAC Proc. Vol. 2009, 42, 1–20. https://doi.org/10.3182/20090706-3-FR-2004.00001.
- 11.
Sarkka, S. Bayesian Filtering and Smoothing; Cambridge University Press: Cambridge, UK, 2013.
- 12.
Schon, T.B.; Lindsten, F.; Dahlin, J.; et al. Sequential Monte Carlo methods for system identification. IFAC-PapersOnLine 2015, 48, 775–786.
- 13.
Dutra, D.A.A. Uncertainty estimation in equality-constrained MAP and maximum likelihood estimation with applications to system identification and state estimation. Automatica 2020, 116, 108935. https://doi.org/10.1016/j.automatica.2020.108935.
- 14.
Shumway, R.H.; Stoffer, D.S. An approach to time series smoothing and forecasting using the EM algorithm. J. Time Ser. Anal. 1982, 3, 253–264.
- 15.
Dempster, A.P.; Laird, N.M.; Rubin, D.B. Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B 1977, 39, 1–22. https://doi.org/10.1111/j.2517-6161.1977.tb01600.x.
- 16.
Schon, T.B.; Wills, A.; Ninness, B. System identification of nonlinear state-space models. Automatica 2011, 47, 39–49. https://doi.org/10.1016/j.automatica.2010.10.013.
- 17.
Kalman, R.E. A New Approach to Linear Filtering and Prediction Problems. J. Basic Eng. 1960, 82, 35–45. https://doi.org/10.1115/1.3662552.
- 18.
Rauch, H.E.; Tung, F.; Striebel, C.T. Maximum Likelihood Estimates of Linear Dynamic Systems. AIAA J. 1965, 3, 1445–1450. https://doi.org/10.2514/3.3166.
- 19.
Julier, S.J.; Uhlmann, J.K. New Extension of the Kalman Filter to Nonlinear Systems. In Proceedings of the Signal Processing, Sensor Fusion, and Target Recognition VI, Orlando, FL, USA, 21–24 April 1997; Volume 3068, pp. 182–193.
- 20.
Arasaratnam, I.; Haykin, S. Cubature Kalman Filters. IEEE Trans. Autom. Control 2009, 54, 1254–1269. https://doi.org/10.1109/TAC.2009.2019800.
- 21.
Kokkala, J.; Solin, A.; Sarkka, S. Sigma-Point Filtering and Smoothing Based Parameter Estimation in Nonlinear Dynamic Systems. J. Adv. Inf. Fusion 2016, 11, 15–30.
- 22.
Arulampalam, M.S.; Maskell, S.; Gordon, N.; et al. A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking. IEEE Trans. Signal Process. 2002, 50, 174–188. https://doi.org/10.1109/78.978374.
- 23.
Doucet, A.; de Freitas, N.; Gordon, N.J. (Eds.) Sequential Monte Carlo Methods in Practice; Springer: New York, NY, USA, 2001.
- 24.
Andrieu, C.; Doucet, A.; Singh, S.S.; et al. Particle methods for change detection, system identification, and control. Proc. IEEE 2004, 92, 423–438.
- 25.
Beal, M.J. Variational Algorithms for Approximate Bayesian Inference. Ph.D. Thesis, University of London, London, UK, 2003.
- 26.
Blei, D.M.; Kucukelbir, A.; McAuliffe, J.D. Variational Inference: A Review for Statisticians. J. Am. Stat. Assoc. 2017, 112, 859–877. https://doi.org/10.1080/01621459.2017.1285773.
- 27.
Lindfors, M.; Chen, T. Regularized LTI system identification in the presence of outliers: A variational EM approach. Automatica 2020, 121, 109152. https://doi.org/10.1016/j.automatica.2020.109152.
- 28.
Tan, L.S.L.; Nott, D.J. Gaussian variational approximation with sparse precision matrices. Stat. Comput. 2018, 28, 259–275. https://doi.org/10.1007/s11222-017-9722-9.
- 29.
Quiroz, M.; Nott, D.J.; Kohn, R. Gaussian variational approximations for high-dimensional state space models. Bayesian Anal. 2023, 18, 989–1016.
- 30.
Barfoot, T.D.; Forbes, J.R.; Yoon, D.J. Exactly Sparse Gaussian Variational Inference with Application to Derivative-Free Batch Nonlinear State Estimation. Int. J. Robot. Res. 2020, 39, 1473–1502. https://doi.org/10.1177/0278364920937608.
- 31.
Courts, J.; Wills, A.G.; Schon, T.B.; et al. Variational system identification for nonlinear state-space models. Automatica 2023, 147, 110687. https://doi.org/10.1016/j.automatica.2022.110687.
- 32.
Dutra, D.A.A. Parameterizations for large-scale variational system identification using unconstrained optimization. Automatica 2025, 173, 112086. https://doi.org/10.1016/j.automatica.2024.112086.
- 33.
Chung, J.; Kastner, K.; Dinh, L.; et al. A Recurrent Latent Variable Model for Sequential Data. In Proceedings of the Advances in Neural Information Processing Systems, Montreal, QC, Canada, 7–12 December 2015; Volume 28.
- 34.
Fraccaro, M.; Sønderby, S.R.K.; Paquet, U.; et al. Sequential Neural Models with Stochastic Layers. In Proceedings of the Advances in Neural Information Processing Systems, Barcelona, Spain, 5–10 December 2016; Volume 29.
- 35.
Krishnan, R.G.; Shalit, U.; Sontag, D. Structured Inference Networks for Nonlinear State Space Models. In Proceedings of the AAAI Conference on Artificial Intelligence, San Francisco, CA, USA, 4–9 February 2017; Volume 31, pp. 2101–2109.
- 36.
Karl, M.; Soelch, M.; Bayer, J.; et al. Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data. In Proceedings of the International Conference on Learning Representations, Toulon, France, 24–26 April 2017.
- 37.
Andersson, C.; Ribeiro, A.H.; Tiels, K.; et al. Deep Convolutional Networks in System Identification. In Proceedings of the 2019 IEEE 58th Conference on Decision and Control (CDC), Nice, France, 11–13 December 2019; pp. 3670–3676.
- 38.
Pillonetto, G.; Aravkin, A.; Gedon, D.; et al. Deep networks for system identification: A survey. Automatica 2025, 171, 111907. https://doi.org/10.1016/j.automatica.2024.111907.
- 39.
Kingma, D.P.; Ba, J. Adam: A Method for Stochastic Optimization. arXiv 2014, arXiv:1412.6980.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Suite 4002 Level 4, 447 Collins Street, Melbourne, Victoria 3000, Australia
General Inquiries: info@sciltp.com