Identifiable Autoregressive Variational Autoencoders for Nonlinear and Nonstationary Spatio-Temporal Blind Source Separation
This work addresses challenges in dimension reduction for spatio-temporal data, offering a method with identifiability guarantees, though it appears incremental by building on recent advances in identifiability results.
The paper tackled the problem of modeling multivariate spatio-temporal data by proposing an identifiable autoregressive variational autoencoder for nonlinear and nonstationary blind source separation, demonstrating its efficacy in simulations and outperforming competitors on air pollution and weather datasets.
The modeling and prediction of multivariate spatio-temporal data involve numerous challenges. Dimension reduction methods can significantly simplify this process, provided that they account for the complex dependencies between variables and across time and space. Nonlinear blind source separation has emerged as a promising approach, particularly following recent advances in identifiability results. Building on these developments, we introduce the identifiable autoregressive variational autoencoder, which ensures the identifiability of latent components consisting of nonstationary autoregressive processes. The blind source separation efficacy of the proposed method is showcased through a simulation study, where it is compared against state-of-the-art methods, and the spatio-temporal prediction performance is evaluated against several competitors on air pollution and weather datasets.