Under the Hood
Despite the fact GMDH Shell is very complex software, we don't make a secret from our core algorithms. You can find details at the GMDH Shell Documentation: Learning algorithms or to have deeper insight of GMDH-family algorithms we suggest to read the Research Report about GMDH method (PDF).
GMDH Shell implements the following GMDH algorithms:
Our Combinatorial algorithm is similar to the classic combinatorial GMDH algorithm called COMBI. You can get its description from the above PDF.
GMDH-type neural networks aka Polynomial Neural Networks
Our Neural-type algorithm is a variation of the classic Multilayer Iterative GMDH algorithm (MIA). The difference between our Neural-type and original MIA algorithm is that there are a number of theoretical improvements introduced by different authors since 1968, so GMDH Shell implements all the features that have proven effective and have been partially included into competing software packages (KnowledgeMiner, NeuroShell). These features are: Active neurons, Neurons with more than 2 inputs, connections that skip layers and bi-criteria neuron selection.