Deep Dynamic Probabilistic Canonical Correlation Analysis
Shiqin Tang, Shujian Yu, Yining Dong, S. Joe Qin
Abstract
This paper presents Deep Dynamic Probabilistic Canonical Correlation Analysis (D<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>PCCA), a model that integrates deep learning with probabilistic modeling to analyze nonlinear dynamical systems. Building on the probabilistic extensions of Canonical Correlation Analysis (CCA), D<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>PCCA captures nonlinear latent dynamics and supports enhancements such as KL annealing for improved convergence and normalizing flows for a more flexible posterior approximation. D<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>PCCA naturally extends to multiple observed variables, making it a versatile tool for encoding prior knowledge about sequential datasets and providing a probabilistic understanding of the system’s dynamics. Experimental validation on real financial datasets demonstrates the effectiveness of D<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>PCCA and its extensions in capturing latent dynamics.
BibTeX
@inproceedings{icassp2025_deepdynamicproba,
title = {Deep Dynamic Probabilistic Canonical Correlation Analysis},
author = {Shiqin Tang and Shujian Yu and Yining Dong and S. Joe Qin},
booktitle = {ICASSP 2025},
year = {2025}
}