Probabilistic Segmentation via Total Variation Regularization

TL;DR

This paper introduces a convex method for probabilistic segmentation of time series data using total variation regularization, enabling efficient training and effective clustering.

Contribution

It presents a novel convex framework for probabilistic time series segmentation that incorporates total variation regularization and efficient optimization techniques.

Findings

Performs as well or better than existing latent variable models.

Easier to train compared to traditional methods.

Effective in real-world segmentation tasks.

Abstract

We present a convex approach to probabilistic segmentation and modeling of time series data. Our approach builds upon recent advances in multivariate total variation regularization, and seeks to learn a separate set of parameters for the distribution over the observations at each time point, but with an additional penalty that encourages the parameters to remain constant over time. We propose efficient optimization methods for solving the resulting (large) optimization problems, and a two-stage procedure for estimating recurring clusters under such models, based upon kernel density estimation. Finally, we show on a number of real-world segmentation tasks, the resulting methods often perform as well or better than existing latent variable models, while being substantially easier to train.