Beyond EM Algorithm on Over-specified Two-Component Location-Scale Gaussian Mixtures

TL;DR

This paper introduces the ELU algorithm for two-component Gaussian mixtures, achieving rapid convergence to the statistical radius with logarithmic iterations, overcoming EM's slow convergence in over-specified models.

Contribution

The paper proposes the ELU algorithm that combines exact scale optimization with exponential gradient steps, providing optimal statistical and computational efficiency for over-specified Gaussian mixtures.

Findings

ELU converges in logarithmic iterations to the statistical radius.

ELU outperforms EM in over-specified models.

Theoretical and empirical validation of ELU's efficiency.

Abstract

The Expectation-Maximization (EM) algorithm has been predominantly used to approximate the maximum likelihood estimation of the location-scale Gaussian mixtures. However, when the models are over-specified, namely, the chosen number of components to fit the data is larger than the unknown true number of components, EM needs a polynomial number of iterations in terms of the sample size to reach the final statistical radius; this is computationally expensive in practice. The slow convergence of EM is due to the missing of the locally strong convexity with respect to the location parameter on the negative population log-likelihood function, i.e., the limit of the negative sample log-likelihood function when the sample size goes to infinity. To efficiently explore the curvature of the negative log-likelihood functions, by specifically considering two-component location-scale Gaussian mixtures, we develop the Exponential Location Update (ELU) algorithm. The idea of the ELU algorithm is that we first obtain the exact optimal solution for the scale parameter and then perform an exponential step-size gradient descent for the location parameter. We demonstrate theoretically and empirically that the ELU iterates converge to the final statistical radius of the models after a logarithmic number of iterations. To the best of our knowledge, it resolves the long-standing open question in the literature about developing an optimization algorithm that has optimal statistical and computational complexities for solving parameter estimation even under some specific settings of the over-specified Gaussian mixture models.