Particle swarm optimization in constrained maximum likelihood estimation a case study

TL;DR

This paper demonstrates that particle swarm optimization, specifically global best and local best variants, is highly effective for solving constrained maximum likelihood estimation problems in bioinformatics, especially when traditional methods fail.

Contribution

The study applies and evaluates PSO algorithms to a bioinformatics problem, highlighting their efficiency in non-differentiable, non-convex optimization scenarios.

Findings

PSO is effective for non-differentiable, non-convex problems

Global best and local best PSO outperform gradient-based methods in this context

PSO provides a practical solution for constrained maximum likelihood estimation in bioinformatics

Abstract

The aim of paper is to apply two types of particle swarm optimization, global best andlocal best PSO to a constrained maximum likelihood estimation problem in pseudotime anal-ysis, a sub-field in bioinformatics. The results have shown that particle swarm optimizationis extremely useful and efficient when the optimization problem is non-differentiable and non-convex so that analytical solution can not be derived and gradient-based methods can not beapplied.