What is a neural network?

What is a neural network? The question is not easy to answer. In the neurosciences, a neuronal network is defined as any number of interconnected neurons which, as part of a nervous system, form a connection that serves a specific function. In computational neuroscience, simplified models of a biological network are also understood in abstract terms.

Instead of less power-intensive arithmetic units, neural networks use a large number of simply constructed arithmetic units. These work in parallel and have learning capabilities. With artificial neural networks, learning happens at by algorithms which modify the network by altering its own neurons and their train connections on the basis of repeatedly presented data. Since neuronal Networks do not simply store data, but associated data through learning, possess the ability to generalise, which means that after successful learning similar data can be associated. In addition, the
Distribution of the data to all neurons in the network has a certain error tolerance. For example, during drunken intoxication neurons die and the brain still works perfectly the next day. However, the distribution of the data also problems, because one cannot recognize at first sight, what a neural network knows, what it can do, or where its faults lie.
The basic idea at the beginning of research into artificial neural networks was to develop a mathematical model for computer use based on the biological model of neurons and their compounds.

Neural Network


In computer science, information technology and robotics, their structures are modelled as artificial neural networks and technically simulated and modified.

In computer science, information technology and robotics, their structures are modelled as artificial neural networks and technically simulated and modified.

The Networking of Neurons
The nervous system of humans and animals consists of nerve cells (neurons) and glial cells as well as an environment. The neurons are linked together by synapses, which can be seen as the connection points or nodes of an interneuronal network. In addition, chemical and electrical exchanges take place between neurons and cells of the neuroglia, in particular oligodendroglia and astroglia, which can change the weighting of signals.

The “circuit technology” of neurons usually knows several inputs and one output. If the sum of the input signals exceeds a certain threshold value, the neuron “fires” (excitation formation): an action potential (AP) is triggered at the axon hill, formed in the initial segment and passed on along the axon (excitation conduction). Action potentials in series are the primary output signals of neurons. These signals can be transmitted to other cells via synapses (excitation transmission). At electrical synapses, the potential changes are passed on in direct contact. At chemical synapses, these are converted into a transmitter quantum as a secondary signal, i.e. transmitted by messenger substances (transmission).

Schematic representation of a simple neuronal network
with divergence: a neuron transmits signals to several other neurons,
and convergence: one neuron receives signals from several others.
Nerve cells are characterised by their cell extensions, which are used to establish contact with individual other cells. As dendrites, they are primarily used to receive signals from other cells, while signals are transmitted to other cells via the neurites, also known as axons in glial cell envelopes.