ABSTRACT
Premature convergence is a persisting problem in evolutionary optimisation, in particular - genetic algorithms. While a number of methods exist to approach this issue, they usually require problem specific calibration or only partially resolve the issue, at best by delaying the premature convergence of an evolving population. Analytical models in biology show that resiliently diverse populations evolve on high-dimensional fitness landscapes with "holey" rather than "rugged" topographies, but the implications for artificial evolutionary systems remain largely unexplored. Here I show how holey fitness landscapes (HFLs) can be incorporated in an evolutionary algorithm and use this approach to investigate the ability of HFLs to maintain genetic diversity in an evolving population. The results indicate that an underlying HFL can counteract premature genetic convergence and sustain diversity. They also suggest that HFL may provide a flexible mechanism for dynamic creation and maintenance of subpopulations that concentrate their evolutionary search in different regions of the solution space. Finally, I discuss on-going work on using the HFL model in optimisation problems.
- S. W. Mahfoud (1995). Niching Methods for Genetic Algorithms. PhD thesis. Urbana Iniversity of Illinois. Google ScholarDigital Library
- K. A. De Jong (1975). Analysis of the behaviour of a class of genetic adaptive systems. PhD thesis University of Michigan. Google ScholarDigital Library
- D. E. Goldberg and J. Richardson (1987). Genetic algorithms with sharing for multimodal function optimization, 2nd Int. Conf. on GAs and their application, pp. 41--49 Lawrence Erlbaum Associates Inc., Mahwah, NJ, USA. Google ScholarDigital Library
- D. Whitley, S. Rana and R. B. Heckendorn (1999). The Island Model Genetic Algorithm: On Separability, Population Size and Convergence. J. of Computing and IT. 7 (1): pp. 33--47.Google Scholar
- J. Liu, Z. Cai and J. Liu (2000). Premature convergence in genetic algorithm: Analysis and prevention based on chaos operator, Procs 3rd W. Congr. Intell. Ctrl and Automation.Google Scholar
- N. Kashtan, E. Noor and U. Alon (2007). Varying environments can speed up evolution. PNAS. 104 (34).Google Scholar
- G. Paperin, S. Sadedin, D. G. Green and A. Dorin (2008). Holey Fitness Landscapes and the Maintenance of Evolutionary Diversity Submitted to 11th International Conference on Simulation and Synthesis of Living Systems (ALife XI).Google Scholar
- S. Gavrilets (2004). Fitness Landscapes and the Origin of Species. Princeton University Press, Princeton / Oxford.Google Scholar
- S. Gavrilets and J. Gravner (1997). Percolation on the Fitness Hypercube and the Evolution of Reproductive Isolation. J. of Th. Bio. 184 (1): pp. 51--64.Google ScholarCross Ref
- G. Paperin, D. G. Green and A. Dorin (2007). Fitness Landscapes in Individual-Based Simulation Models of Adaptive Radiation. In T. D. Pham and X. Zhou (eds.), 2007 Int. Symposium on Computational Models for Life Science .Google Scholar
- G. Paperin, D. G. Green, S. Sadedin and T. G. Leishman (2007). A Dual Phase Evolution model of adaptive radiation in landscapes. In M. Randall, H. A. Abbass and J. Wiles (eds.), The 3rd Australian Conference on Artificial Life (ACAL'07), Springer. Google ScholarDigital Library
- S. Gavrilets (2003). Models of Speciation: What have we learned in 40 years? Evolution. 57 (10): pp. 2197--2215.Google ScholarCross Ref
- M. E. J. Newman and R. M. Ziff (2000). Efficient Monte Carlo Algorithm and High-Precision Results for Percolation. Physical Review Letters. 85 (19): pp. 4104--4107.Google ScholarCross Ref
- G. R. Grimmett (1999). Percolation. Springer.Google Scholar
- S. Gavrilets and A. Vose (2005). Dynamic patterns of adaptive radiation. PNAS. 102 (50): pp. 18040--18045.Google ScholarCross Ref
- T. Ohta and M. Kimura (1973). A model of mutation appropriate to estimate the number of electrophoretically detectable alleles in a population. Gen. Res. 22 (2): pp. 201--204.Google ScholarCross Ref
- S. Van Dongen (2000). Graph Clustering by Flow Simulation. PhD thesis University of Utrecht. Utrecht.Google Scholar
- R. R. Hudson, M. Slatkin and W. P. Maddison (1992). Estimation of Levels of Gene Flow From DNA Sequence Data. Genetics. 132 (2): pp. 583--589.Google Scholar
- LiveGraph - a framework for real-time data visualisation, analysis and logging. Retrieved on 01.03.2008 from: http://www.live-graph.org.Google Scholar
Index Terms
- Using holey fitness landscapes to counteract premature convergence in evolutionary algorithms
Recommendations
Maintaining Genetic Diversity in Multimodal Evolutionary Algorithms using Population Injection
GECCO '16 Companion: Proceedings of the 2016 on Genetic and Evolutionary Computation Conference CompanionIn this paper, we present a computationally inexpensive method for maintaining genetic diversity in evolutionary algorithms using population injection. As opposed to other methods, e.g., cellular EAs, population injection does not require any ...
A synergistic approach for evolutionary optimization
GECCO '08: Proceedings of the 10th annual conference companion on Genetic and evolutionary computationOne of the major causes of premature convergence in Evolutionary Algorithm (EA) is loss of population diversity, which pushes the search space to a homogeneous or a near-homogeneous configuration. In particular, this can be a more complicated issue in ...
Fitness landscapes and graphs: multimodularity, ruggedness and neutrality
GECCO '13 Companion: Proceedings of the 15th annual conference companion on Genetic and evolutionary computationOne of the most commonly-used metaphors to describe the process of heuristic search methods in solving combinatorial optimization problems is that of a fitness landscape. The landscape metaphor appears most commonly in work related to evolutionary ...
Comments