Dynamical Wasserstein Barycenters for Time-series Modeling
arXiv:2110.06741
Abstract
Many time series can be modeled as a sequence of segments representing high-level discrete states, such as running and walking in a human activity application. Flexible models should describe the system state and observations in stationary "pure-state" periods as well as transition periods between adjacent segments, such as a gradual slowdown between running and walking. However, most prior work assumes instantaneous transitions between pure discrete states. We propose a dynamical Wasserstein barycentric (DWB) model that estimates the system state over time as well as the data-generating distributions of pure states in an unsupervised manner. Our model assumes each pure state generates data from a multivariate normal distribution, and characterizes transitions between states via displacement-interpolation specified by the Wasserstein barycenter. The system state is represented by a barycentric weight vector which evolves over time via a random walk on the simplex. Parameter learning leverages the natural Riemannian geometry of Gaussian distributions under the Wasserstein distance, which leads to improved convergence speeds. Experiments on several human activity datasets show that our proposed DWB model accurately learns the generating distribution of pure states while improving state estimation for transition periods compared to the commonly used linear interpolation mixture models.
To appear at Neurips 2021
References in corpus (8)
- Adam: A Method for Stochastic Optimization
- Statistical Aspects of Wasserstein Distances
- Riemannian Geometry of Symmetric Positive Definite Matrices via Cholesky Decomposition
- Wasserstein Riemannian Geometry of Positive Definite Matrices
- A fixed-point approach to barycenters in Wasserstein space
- Optimal transport for Gaussian mixture models
- Manifold Optimization for Gaussian Mixture Models
- On a class of random walks in simplexes