Competition on Dynamic Optimization Problems Generated by Generalized Moving Peaks Benchmark (GMPB)
Danial Yazdani, Michalis Mavrovouniotis, Changhe Li, Guoyu Chen, Wenjian Luo, Mohammad Nabi Omidvar, Juergen Branke, Shengxiang Yang, Xin Yao
TL;DR
The paper introduces GMPB, a generalized and highly configurable benchmark for dynamic optimization that extends the Moving Peaks Benchmark to multiple evolving peaks. It defines a max-over-components landscape with rotating, irregular, and interactively evolving peaks, enabling nonseparable and potentially ill-conditioned problem instances. The authors detail parameter settings, environment dynamics, and two performance indicators (EO and BBC) to rigorously evaluate adaptive algorithms, and situate GMPB within the EDOLAB MATLAB platform for broad accessibility. This work provides a versatile, comparable testbed for assessing online optimization methods in dynamic and morphologically varied landscapes with practical tooling for experimentation and visualization.
Abstract
The Generalized Moving Peaks Benchmark (GMPB) is a tool for generating continuous dynamic optimization problem instances with controllable dynamic and morphological characteristics. GMPB has been used in recent Competitions on Dynamic Optimization at prestigious conferences, such as the IEEE Congress on Evolutionary Computation (CEC). This dynamic benchmark generator can create a wide variety of landscapes, ranging from simple unimodal to highly complex multimodal configurations and from symmetric to asymmetric forms. It also supports diverse surface textures, from smooth to highly irregular, and can generate varying levels of variable interaction and conditioning. This document provides an overview of GMPB, emphasizing how its parameters can be adjusted to produce landscapes with customizable characteristics. The MATLAB implementation of GMPB is available on the EDOLAB Platform.