End-to-End Identifiable and Consistent Recurrent Switching Dynamical Systems
For researchers working on identifiable deep generative models for sequential data with regime-switching, this paper provides both theoretical guarantees and a practical estimator that outperforms existing VAE-based methods.
This work establishes identifiability for a broad class of recurrent nonlinear switching dynamical systems under flexible assumptions and introduces ΩSDS, a flow-based estimator using expectation-maximisation for exact likelihood optimization. Empirically, ΩSDS achieves improved disentanglement and more accurate forecasting compared to VAE-based estimators.
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 $Ω$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 $Ω$SDS achieves improved disentanglement compared to VAE-based estimators and more accurate forecasting of underlying dynamics.