Abstract | ||
---|---|---|
In this paper we present application of genetic programming (GP) [1] to evolution of indirect encoding of neural network weights. We compare usage of original HyperNEAT algorithm with our implementation, in which we replaced the underlying NEAT with genetic programming. The algorithm was named HyperGP. The evolved neural networks were used as controllers of autonomous mobile agents (robots) in simulation. The agents were trained to drive with maximum average speed. This forces them to learn how to drive on roads and avoid collisions. The genetic programming lacking the NEAT complexification property shows better exploration ability and tends to generate more complex solutions in fewer generations. On the other hand, the basic genetic programming generates quite complex functions for weights generation. Both approaches generate neural controllers with similar abilities. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1007/978-3-642-04921-7_25 | ICANNGA |
Keywords | Field | DocType |
neural controller,underlying neat,neural network,complex solution,complex function,neat complexification property,genetic programming,neural network weight,basic genetic programming,original hyperneat algorithm,mobile agent | Computer science,Mobile agent,HyperNEAT,Recurrent neural network,Genetic programming,Artificial intelligence,Neural network controller,Robot,Artificial neural network,Machine learning,Encoding (memory) | Conference |
Volume | ISSN | ISBN |
5495.0 | 0302-9743 | 3-642-04920-6 |
Citations | PageRank | References |
13 | 1.09 | 10 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zdeněk Buk | 1 | 16 | 1.61 |
Jan Koutník | 2 | 552 | 36.31 |
Miroslav Šnorek | 3 | 49 | 6.41 |