Abstract | ||
---|---|---|
In this paper artificial regulatory networks (ARN) are evolved to match the dynamics of test functions. The ARNs are based on a genome representation generated by a duplication / divergence process. By creating a mapping between the protein concentrations created by gene excitation and inhibition to an output function, the network can be evolved to match output functions such as sinusoids, exponentials and sigmoids. This shows that the dynamics of an ARN may be evolved and thus may be suitable as a method for generating arbitrary time-series for function optimization. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1007/978-3-540-30217-9_58 | Lecture Notes in Computer Science |
Keywords | Field | DocType |
time series | Exponential function,Network motif,Computer science,Biomimetics,Recurrent neural network,Algorithm,Function optimization,Network model,Genetic program | Conference |
Volume | ISSN | Citations |
3242 | 0302-9743 | 18 |
PageRank | References | Authors |
1.13 | 16 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
P. Dwight Kuo | 1 | 42 | 3.92 |
André Leier | 2 | 197 | 19.87 |
Wolfgang Banzhaf | 3 | 2627 | 367.13 |