The Influence of Learning in the Evolution of Busy Beavers
The goal of this research is to study how individual learning interacts with an evolutionary algorithm in its search for good candidates for the Busy Beaver problem. Two learning models, designed to act as local search procedures, are proposed. Experimental results show that local search methods that are able to perform several modifications in the structure of an individual in each learning step provide an important advantage. Some insight about the role that evolution and learning play during search is also presented.
Unable to display preview. Download preview PDF.
- 1.Belew, R. and Mitchell, M. (1996). Adaptive Individuals in Evolving Populations: Models and Algorithms, Santa Fe Institute in the Sciences of Complexity, Vol. XXVI, Reading, MA: Addison-Wesley.Google Scholar
- 2.Pereira, F.B., Machado, P., Costa, E., Cardoso, A., Rodriguez, A. Santana, R., and Soto, M. (2000). Too Busy to Learn. Proceedings of the Congress on Evolutionary Computation (CEC-2000), pp. 720–727.Google Scholar
- 3.Sasaki, T. and Tokoro, M. (1999). Adaptation under Changing Environments with Various Rates of Inheritance of Acquired Characters: Comparison Between Darwinian and Lamarckian Evolution. In McKay, B., Yao, X., Newton, C.S., Kim, J.H. and Furuhashi, T. (Eds.), Proceedings of 2 nd Asia-Pacific Conference on Simulated Evolution and Learning (SEAL-98).Google Scholar
- 4.Whitley, D., Gordon, S. and Mathias, K. (1994). Lamarckian Evolution, the Baldwin Effect, and Function Optimization. In Davidor, Y. Schwefel, H.P. and Manner, R. (Eds.) Parallel Problem Solving from Nature (PPSN-III), pp. 6–15.Google Scholar
- 5.Corne, D, Glover, F. and Dorigo, M. (1999). New Ideas in Optimization. McGraw-Hill.Google Scholar
- 7.Jones, T. and Rawlins, G. (1993). Reverse Hillclimbing, Genetic Algorithms, and the Busy Beaver Problem. In Forrest, S. (Ed.). Proceedings of the 5 th International Conference on Genetic Algorithms (ICGA-93), pp.70–75, San Mateo, CA, Morgan Kaufmann.Google Scholar
- 8.Machado, P., Pereira, F.B, Cardoso, A., Costa, E. (1999). Busy Beaver-The Influence of Representation, In Poli, R., Nordin, P. Langdon, W. and Fogarty, T. (Eds.). Proceedings of the Second European Workshop in Genetic Programming (EuroGP-99).Google Scholar
- 10.Pereira, F.B. and Costa, E. (1997). The Influence of Learning in the Optimization of Royal Road Functions, Proceedings of the 3 rd International Mendel Conference on Genetic algorithms, Optimization Problems, Fuzzy Logic, Neural Networks and Rough Sets (Mendel’97), pp 244–249.Google Scholar
- 11.Boolos, G., and Jeffrey, R. (1995). Computability and Logic, Cambridge University Press.Google Scholar
- 12.Ackley, D. and Littman, M. (1994). A Case for Lamarckian Evolution. In Langton, C. (Ed.), Artificial Life III, pp. 3–10, Addison-Wesley.Google Scholar
- 13.Imada, A. and Araki, K. (1996). Lamarckian Evolution of Associative Memory. In Proceedings of the Third International Conference on Evolutionary Computation (ICEC-96), pp.676–680.Google Scholar
- 14.Pereira, F.B., Machado, P., Costa, E. and Cardoso A. (1999). Graph Based Crossover-A Case Study with the Busy Beaver Problem. In Banzhaf, W., Daida, J., Eiben, A.E., Garzon, M.H., Honavar, V., Jakiela, M., & Smith, R.E. (Eds.). GECCO-99: Proceedings of the Genetic and Evolutionary Computation Conference, pp. 1149–1155, Morgan Kaufmann.Google Scholar
- 15.Lally, A., Reineke, J. and Weader, J. (1997). An Abstract Representation of Busy Beaver Candidate Turing Machines, Technical Report, Van Gogh Group, Rensselaer Polytechnic Institute.Google Scholar