Neural Networks

neuronal networks:
It is an data processing paradigm that is inspired by the biological neurons i.e. the way in which data processing takes go down in a human brain with the help of the neurons.
Neural networks consists of following things:
1) Feed forward mechanism: this means that the call attention quite a little be transmitted one way and i.e. input signal to rig only. There is no feedback (loops) i.e. the output of all layer does not affect that same layer. Feed-forward ANNs tend to be straight forward networks that associate inputs with outputs. They are extensively use in pattern recognition. This type of organization is also referred to as bottom-up or top-down.
2) Layers:
a) Input layer: it represents the raw information that is fed into the network
b) Hidden layer: its drill depends on the bodily do of the input layer and the weights on the connection between the input and the hidden layer
c) Output layer: its activity depends on the activity of the output layer and the weights on the connection between the output layer and the hidden layer
3) Weights: The connection sets whether one unit can influence other unit or not and weights determine the extent of this influence.
4) Transfer function: it is the input output function specified for the units. A transfer function can be: (?1+x1w1+x2w2…..



+ ?2+x3w3+……….)
a) Linear: output is directly proportional to the total weighted output
b) Threshold: output is set at one of the two levels, depending on whether the total input is greater or less than the threshold value
c) Sigmoid: output varies infinitely but not linearly as the input changes.
5) mistake: it is the prediction or forecasting error. It is the error between the veritable and the desired output.
6) Learning rate: the rate or fixedness at which the network learns to recognize the pattern is referred to as the attainment rate of the network.
Back propagation of error:
In straddle to train the neural networks we change the...If you want to get a full essay, order it on our website: Ordercustompaper.com



If you want to get a full essay, wisit our page: write my paper