A Neuro-Genetic Framework for Pattern Recognition in Complex Systems
Stefania Bandini, Leonardo Vanneschi, Andrew Wuensche, Alessandro Bahgat Shehata · BOA (University of Milano-Bicocca) · 2008
This paper presents a general framework to automatically generate rules that produce given spatial patterns in complex systems. The proposed framework integrates Genetic Algorithms with Artificial Neural Networks or Support Vector Machines. Here, it is tested on a well known 3-values, 6-neighbors, k-totalistic cellular automata rule called the burning paper rule. Results are encouraging and should pave the way for the use of the proposed framework in real-life complex systems models.