What I learned
Artificial neural networks are computing systems vaguely inspired by the biological neural networks that constitute animal brains. They are based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain. They are designed to form a Machine Learning system that can learn and perform tasks such as discrimination and classification.
ANN are by no means supposed to constitute a realistic modelization of the way the human brain is working.
An artificial neuron is an abstract computer structure which acts as a basic processing unit. It receives N input signals and transforms them into a single signal using a weighted sum, then it uses an activation function to fire the output signal(s).
All ANN share the following features:
- Several inputs coming either from ‘outside’ or from other processing units (dendrites);
- ‘Weights’ indicating how an input signal influence the processing unit that receives it (this is the frequency & nature of the electric signal received via the synapses);
- A function which sums up all the inputs (the addition of all input signals in the neuron);
- A system to compute a ceiling value. If the sum of the inputs exceeds that ceiling then the signal is transmitted, otherwise, the signal is not transmitted
- An outgoing signal (the signal sending to the outside through the axon);