Flat-topped Probability Density Functions for Mixture Models

03/31/2022

∙

by   Osamu Fujita, et al.

∙

0

∙

share

This paper investigates probability density functions (PDFs) that are continuous everywhere, nearly uniform around the mode of distribution, and adaptable to a variety of distribution shapes ranging from bell-shaped to rectangular. From the viewpoint of computational tractability, the PDF based on the Fermi-Dirac or logistic function is advantageous in estimating its shape parameters. The most appropriate PDF for n-variate distribution is of the form: p(𝐱)∝[cosh([(𝐱-𝐦)^𝖳Σ^-1(𝐱-𝐦)]^n/2)+cosh(r^n)]^-1 where 𝐱,𝐦∈ℝ^n, Σ is an n× n positive definite matrix, and r>0 is a shape parameter. The flat-topped PDFs can be used as a component of mixture models in machine learning to improve goodness of fit and make a model as simple as possible.