A general solver to the elliptical mixture model through an approximate Wasserstein manifold

06/09/2019
by   Shengxi Li, et al.
0

This paper studies the problem of estimation for general finite mixture models, with a particular focus on the elliptical mixture models (EMMs). Instead of using the widely adopted Kullback-Leibler divergence, we provide a stable solution to the EMMs that is robust to initialisations and attains superior local optimum by adaptively optimising along a manifold of an approximate Wasserstein distance. More specifically, we first summarise computable and identifiable EMMs, in order to identify the optimisation problem. Due to a probability constraint, solving this problem is cumbersome and unstable, especially under the Wasserstein distance. We thus resort to an efficient optimisation on a statistical manifold defined under an approximate Wasserstein distance, which allows for explicit metrics and operations. This is shown to significantly stabilise and improve the EMM estimations. We also propose an adaptive method to further accelerate the convergence. Experimental results demonstrate excellent performances of the proposed solver.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/28/2021

Schema matching using Gaussian mixture models with Wasserstein distance

Gaussian mixture models find their place as a powerful tool, mostly in t...
research
09/15/2011

Convergence of latent mixing measures in finite and infinite mixture models

This paper studies convergence behavior of latent mixing measures that a...
research
05/21/2018

A universal framework for learning based on the elliptical mixture model (EMM)

An increasing prominence of unbalanced and noisy data highlights the imp...
research
09/15/2023

Augmented quantization: a general approach to mixture models

The investigation of mixture models is a key to understand and visualize...
research
08/13/2018

The Gromov-Wasserstein distance between networks and stable network invariants

We define a metric---the Network Gromov-Wasserstein distance---on weight...
research
02/01/2019

Normalized Wasserstein Distance for Mixture Distributions with Applications in Adversarial Learning and Domain Adaptation

Understanding proper distance measures between distributions is at the c...
research
06/16/2021

Regularization of Mixture Models for Robust Principal Graph Learning

A regularized version of Mixture Models is proposed to learn a principal...

Please sign up or login with your details

Forgot password? Click here to reset