This could increase training time. The attributes with the largest range would have an undue influence during the operation of the learning algorithm. If you had a practical limit on the training time it could lead to a sub-optimal network.

A neuron model is a mathematical model of the behaviour of a single neuron in a biological nervous system. The most important neuron model is the so called perceptron. The perceptron neuron model receives information in the form of numerical inputs. This information is then combined with a set of parameters to produce a message in the form of a single numerical output.

Most neural networks, even biological neural networks, exhibit a layered structure. In this work layers are the basis to determine the architecture of a neural network. A layer of perceptrons takes a set of inputs in order to produce a set of outputs.

A deep neural network is built up by organizing layers of perceptrons in a network architecture. In this way, the architecture of a network refers to the number of layers, their arrangement and connectivity. The characteristic network architecture in OpenNN is the so called feed-forward architecture. The multilayer perceptron can then be defined as a network architecture of perceptron layers. This neural network represents a parameterized function of several variables with very good approximation properties.

In order to solve practical applications, different extensions must be added to the multilayer perceptron. Some of them include scaling, unscaling, bounding, probabilistic or conditions layers. Therefore, the neural network in OpenNN is composed by a mutlilayer perceptron plus some additional layers

Some papers I've read suggest that tanh is superior to logistic for training time and model fit. However, in my experience, tanh ANNs have a tendency to diverge (weights grow forever) when inputs and outputs are not normalized. By contrast, I've found logistic networks to be more robust to unnormalized values.

no convergences