Finite Sample Analysis of Tensor Decomposition for Learning Mixtures of Linear Systems

TL;DR

This paper introduces a tensor decomposition-based method for learning mixtures of linear dynamical systems from input-output data, improving estimation accuracy by leveraging entire trajectories and providing sample complexity bounds.

Contribution

It proposes a novel moment-based estimator using tensor decomposition that enhances existing methods for MLDS by utilizing full trajectory data and analyzing sample complexity.

Findings

Estimator outperforms previous methods in simulations.

Provides explicit sample complexity bounds in noisy settings.

Utilizes entire trajectory length for improved parameter estimation.

Abstract

We study the problem of learning mixtures of linear dynamical systems (MLDS) from input-output data. The mixture setting allows us to leverage observations from related dynamical systems to improve the estimation of individual models. Building on spectral methods for mixtures of linear regressions, we propose a moment-based estimator that uses tensor decomposition to estimate the impulse response parameters of the mixture models. The estimator improves upon existing tensor decomposition approaches for MLDS by utilizing the entire length of the observed trajectories. We provide sample complexity bounds for estimating MLDS in the presence of noise, in terms of both the number of trajectories $N$ and the trajectory length $T$, and demonstrate the performance of the estimator through simulations.