Java Implementation of a Parameter-less Evolutionary Portfolio

TL;DR

This paper presents a Java-based implementation of a portfolio of parameter-less evolutionary algorithms that adaptively select algorithms during runtime, eliminating the need for prior parameter tuning and effectively solving diverse problems.

Contribution

It introduces a novel portfolio approach that combines multiple parameter-less evolutionary algorithms with adaptive selection based on performance criteria.

Findings

The portfolio can solve various problem classes without prior parameter tuning.

The approach maintains acceptable computational effort.

Initial experiments demonstrate effectiveness across problem types.

Abstract

The Java implementation of a portfolio of parameter-less evolutionary algorithms is presented. The Parameter-less Evolutionary Portfolio implements a heuristic that performs adaptive selection of parameter-less evolutionary algorithms in accordance with performance criteria that are measured during running time. At present time, the portfolio includes three parameter-less evolutionary algorithms: Parameter-less Univariate Marginal Distribution Algorithm, Parameter-less Extended Compact Genetic Algorithm, and Parameter-less Hierarchical Bayesian Optimization Algorithm. Initial experiments showed that the parameter-less portfolio can solve various classes of problems without the need for any prior parameter setting technique and with an increase in computational effort that can be considered acceptable.