- All Implemented Interfaces:
- java.io.Serializable
public class Substrate
extends java.lang.Object
implements java.io.Serializable
The substrate defines the structure of the produced HyperNEAT network.
A substrate is made up of nodes and links. A node has a location that is an
n-dimensional coordinate. Nodes are grouped into input and output clusters.
There can also be hidden neurons between these two.
A HyperNEAT network works by training a CPPN that produces the actual
resulting NEAT network. The size of the substrate can then be adjusted to
create larger networks than what the HyperNEAT network was originally trained
with.
-----------------------------------------------------------------------------
http://www.cs.ucf.edu/~kstanley/ Encog's NEAT implementation was drawn from
the following three Journal Articles. For more complete BibTeX sources, see
NEATNetwork.java.
Evolving Neural Networks Through Augmenting Topologies
Generating Large-Scale Neural Networks Through Discovering Geometric
Regularities
Automatic feature selection in neuroevolution
- See Also:
- Serialized Form