Ali,

you have chosen a complex topic to discuss which you might want t break down into several questions. To begin, the field of neural networks started wit the goal of finding appropriate models for the human brain. In the beginning models of McCulloch-Pitts, Hebb, Adaptive Resonance, etc. the goal was to try to find suitable models of neural behavior. The beginning divergence came with back propagation which Has been debated on whether there is any biological justification for the model (there are arguments in favor and against it). Also the drive to learn boolean functions turned the biological problem into a computational problem. This situation was exacerbated by the rise of support vector machines in the 90's in which research into neural networks had to compete against other paradigms. Since then the field has diverged more into optimization problems.

The new deep learning paradigm with restricted Boltzmann machines take its inspiration from Hopfield's original work but since those days it has changed a lot and while there is an initial inspiration of deep learning as do biological models t, the optimization techniques do not have any such basis(at least the mainstream ones). Convolutional Networks do have a biological underpinning, but their implementation relies on non biological foundations.

In summary, There are some biological elements that are worth exploring under the artificial neural network models(such as image abstraction models in convolutional networks), but you have to be explicit on which elements you are taking into account.

If you want we could discuss further specific models and evaluate their biological worth

Please see the attached file.

Hi. A neural network is a powerful data modeling tool that is able to capture and represent complex input/output relationships. Imagine the power of the machine which has the abilities of both computers and humans. It would be the most remarkable thing ever. A neural network usually involves a large number of processors operating in parallel, each with its own small sphere of knowledge and access to data in its local memory. The computing world has a lot to gain from neural networks. Their ability to learn by example makes them very flexible and powerful. They are also very well suited for real time systems because of their fast response and computational times which are due to their parallel architecture

We are more and more aware that the mind of a creature created by the Creator is an incredibly magical one in recognizing complex objects. Neural net is a statistical approach of a collection of experiences to be automated with a computer processor system in experimenting to recognize objects (which are generally still very simple).