Runtime analysis of competitive co-evolutionary algorithms for maximin optimisation of a bilinear function
Author:
Open Access
- 8 July 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1408-1416
Abstract
Co-evolutionary algorithms have a wide range of applications, such as in hardware design, evolution of strategies for board games, and patching software bugs. However, these algorithms are poorly understood and applications are often limited by pathological behaviour, such as loss of gradient, relative over-generalisation, and mediocre objective stasis. It is an open challenge to develop a theory that can predict when co-evolutionary algorithms find solutions efficiently and reliably. This paper provides a first step in developing runtime analysis for population-based competitive co-evolutionary algorithms. We provide a mathematical framework for describing and reasoning about the performance of co-evolutionary processes. An example application of the framework shows a scenario where a simple coevolutionary algorithm obtains a solution in polynomial expected time. Finally, we describe settings where the co-evolutionary algorithm needs exponential time with overwhelmingly high probability to obtain a solution.Keywords
All Related Versions
Funding Information
- Engineering and Physical Sciences Research Council (EP/V025562/1))
This publication has 19 references indexed in Scilit:
- Coevolutionary Principles2012
- Fitness-levels for non-elitist populations2011
- Negative Drift in Populations2010
- Error Thresholds in Genetic AlgorithmsEvolutionary Computation, 2006
- The Cooperative Coevolutionary (1+1) EAEvolutionary Computation, 2004
- Upper and Lower Bounds for Randomized Search Heuristics in Black-Box OptimizationTheory of Computing Systems, 2004
- A New Look at Solving Minimax Problems with Coevolutionary Genetic Algorithms2003
- Three Generations of Automatically Designed RobotsArtificial Life, 2001
- Cooperative Coevolution: An Architecture for Evolving Coadapted SubcomponentsEvolutionary Computation, 2000
- Co-evolving parasites improve simulated evolution as an optimization procedurePhysica D: Nonlinear Phenomena, 1990