Year: 2020
Type: preprint
Abstract: In this letter, we revisit the problem of maximum likelihood estimation (MLE) of parameters of Gaussian Mixture Model (GMM) and show a new derivation for its parameters. The new derivation, unlike the classical approach employing the technique of expectation-maximization (EM), is straightforward and doesn't invoke any hidden or latent variables and calculation of the conditional density function. The new derivation is based on the approach of minorization-maximization and involves finding a tighter lower bound of the log-likelihood criterion. The update steps of the parameters, obtained via the new derivation, are same as the update steps obtained via the classical EM algorithm.
Source: arXiv (Cornell University)
Authors: Sahu, Nitesh, Babu, Prabhu
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3000731331