Evolving adaptive recurrent networks with Developmental Symbolic Encoding

• Autorzy: Suchorzewski M.
• Języki publikacji: EN

Abstrakty

EN

In this paper we are concerned with evolutionary synthesis of recurrent networks capable of learning in the environment. First, we define the model of network we aim to evolve, which is weightless recurrent network of basic arithmetic nodes. Next, we propose a developmental genetic representation for the network, along with some genetic operators for it. The representation bears some important characteristics such as closure and completeness. Most notably, however, it features modularity and scalability, which we demonstrate on a parity problem. Finally, we evolve the network capable of successful learning in some narrow problem domain. The result shows, that for a given problem domain, evolutionary approach may produce networks performing better than generic neural networks.

Słowa kluczowe

EN

evolutionary algorithms; learning; development; intelligent agents; recurrent network

PL

algorytmy ewolucyjne; uczenie się; rozwój; inteligentni agenci; sieci rekurencyjne

• Wydawca: Komisja Informatyki Polskiej Akademii Nauk, Oddział w Gdańsku
• Czasopismo: Metody Informatyki Stosowanej
• Rocznik: 2010
• Tom: nr 1 (22)
• Strony: 175--187
• Opis fizyczny: Bibliogr. 13 poz., rys.

Twórcy

autor

Suchorzewski M.

• West Pomeranian University of Technology, Szczecin Faculty of Computer Science and Information Technology

Bibliografia

• [1] J. Koza, M. Keane, M. Streeter. What’s AI done for me lately? Genetic programming’s human-competitive results. IEEE Intelligent Systems, pp. 25–31, 2003.
• [2] K. Balakrishnan, V. Honavar. Evolutionary and Neural Synthesis of Intelligent Agents. Advances in the evolutionary synthesis of intelligent agents, pp. 1–27. The MIT Press, 2001.
• [3] D. Floreano, C. Mattiussi. Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies.The MIT Press, 2008.
• [4] X. Yao. Evolving Artificial Neural Networks. Proceedings of the IEEE, 87(9):1423–1447, 1999.
• [5] R. Miikkulainen. Evolving neural networks. Proceedings of the 2007 GECCO conference kompanion on Genetic and evolutionary computation, pp. 3415–3434. ACM, 2007.
• [6] D. Floreano, P. D¨urr, C. Mattiussi. Neuroevolution: from architectures to learning. Evolutionary Intelligence, 1(1):47–62, 2008.
• [7] D. Chalmers. The evolution of learning: An experiment in genetic connectionism. Proceedings of the 1990 connectionist models summer school, pp. 81–90. Citeseer, 1990.
• [8] H. Kitano. Designing neural networks using genetic algorithms with graph generation system. Complex Systems, 4(4):461–476, 1990.
• [9] F. Gruau. Neural Network Synthesis using Cellular Encoding and the Genetic Algorithm. PhD thesis, Ecole Normale Suprieure de Lyon, France, 1994.
• [10] F. Gruau, K. Quatramaran. Cellular encoding for interactive evolutionary robotics. Fourth European Conference on Artificial Life, pp. 368–377. The MIT Press, 1997.
• [11] E. Boers, I. Sprinkhuizen-Kuyper. Combined biological metaphors. Advances in the evolutionary synthesis of intelligent agents, pp. 153–183. The MIT Press, 2001.
• [12] K. Stanley, R. Miikkulainen. A taxonomy for artificial embryogeny. Artificial Life, 9(2):93–130, 2003.
• [13] R. Poli, W. Langdon, N. McPhee. A Field Guide to Genetic Programming. Lulu Press, 2008.