Nonparametric and Regularized Dynamical Wasserstein Barycenters for Sequential Observations

TL;DR

This paper extends the dynamical Wasserstein barycenter model for sequential data by relaxing distribution assumptions, introducing regularization for better parameter identification, and demonstrating its effectiveness in human activity segmentation.

Contribution

It proposes a nonparametric, regularized approach to dynamical Wasserstein barycenters for sequential observations, improving model flexibility and robustness.

Findings

Effective segmentation of simulated data

Successful application to real human activity data

Regularization improves parameter identifiability

Abstract

We consider probabilistic models for sequential observations which exhibit gradual transitions among a finite number of states. We are particularly motivated by applications such as human activity analysis where observed accelerometer time series contains segments representing distinct activities, which we call pure states, as well as periods characterized by continuous transition among these pure states. To capture this transitory behavior, the dynamical Wasserstein barycenter (DWB) model of Cheng et al. in 2021 [1] associates with each pure state a data-generating distribution and models the continuous transitions among these states as a Wasserstein barycenter of these distributions with dynamically evolving weights. Focusing on the univariate case where Wasserstein distances and barycenters can be computed in closed form, we extend [1] specifically relaxing the parameterization of the pure states as Gaussian distributions. We highlight issues related to the uniqueness in identifying the model parameters as well as uncertainties induced when estimating a dynamically evolving distribution from a limited number of samples. To ameliorate non-uniqueness, we introduce regularization that imposes temporal smoothness on the dynamics of the barycentric weights. A quantile-based approximation of the pure state distributions yields a finite dimensional estimation problem which we numerically solve using cyclic descent alternating between updates to the pure-state quantile functions and the barycentric weights. We demonstrate the utility of the proposed algorithm in segmenting both simulated and real world human activity time series.