End-to-End Identifiable and Consistent Recurrent Switching Dynamical Systems

Carles Balsells-Rodas, Zhengrui Xiang, Xavier Sumba, Yingzhen Li

Keywords: recurrent switching dynamical systems, nonlinear ICA, sequential representation learning, deep generative models, normalizing flows, identifiability, consistency, disentanglement, time-series model

TL;DR: We establish identifiability results for recurrent switching nonlinear dynamical systems, and propose an exact likelihood-based estimator.

Abstract: Learning identifiable representations in deep generative models remains a fundamental challenge, particularly for sequential data with regime-switching dynamics. Existing approaches establish identifiability under restrictive assumptions, such as stationarity or limited emission models, and typically rely on variational autoencoder (VAE) estimators, which introduce approximation gaps that limit the recovery of the latent structure. In this work, we address both the theoretical and practical limitations of this setting. First, we establish identifiability of a broad class of recurrent nonlinear switching dynamical systems under flexible assumptions, significantly extending prior results. Second, we introduce $\Omega$SDS, a flow-based estimator that enables exact likelihood optimization using expectation-maximisation. Through empirical validation on both synthetic and real-world data, our results demonstrate that $\Omega$SDS achieves improved disentanglement compared to VAE-based estimators and more accurate dynamics forecasting.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 90