Artificial Neural Networks in Machine learning

Artificial Neural Networks(ANN) are mathematical models inspired by biological neural network (Brain).
ANN are made of several neurons, arranged in some form of pattern, which is know as neural network architecture. These neurons are arranged in different layers, each layer can contain 1 or more neurons.

Neural network are either feed forward or feedback.

Feed Forward Neural Networks

In feed forward certain layer is only connected to next layer,

Feed Back Neural Networks:

In feed back neural network certain layer can also have connect with previous layer.

Neurons

Neurons in neural network contain weights, which are multiplied by inputs to produce the output. Different inputs are multiplied to different weights, result of all these input to a certain neuron are added together to produce the final result.

z = X0W0+X1W1+ ..... XnWn 

This formula can also be represented in the form of matrix multiplication.

Single Neuron Representation in Matrix Form

 You can also have multiple neurons, which in formula mean more rows of weights in matrix.

Single Layer Representation of Neural Network

Out of the matrix will be used as an input for the next layer in the same format.