What is a Perceptron ?
Perceptron is a learning algorithm used in machine learning which is used
to do binary classification and decide the input category and provide output
according to the input given.
Single Layer Perceptron :
It is the simplest type of Artificial Neural Network.It is calculated by computing
the sum of all input vector multiplied by corresponding weights.The value which
is displayed in the output will be the value of the Activation function.
Single Layer Perceptron Sums up all the weighted values and and if the sum is
above the predetermined value then the SLP is said to be activated ie Output =1.
Multiple Layer Perceptron - Back Propagation Algorithm.
It is a perceptron that teams up with additional perceptrons stacked up in several layers.
Each perceptron in the first Layer sends an output to the second layer and then the
Second layer sends the output to the third layer and finally like this to the final layer.
For each signal the perceptron uses different weights. every line going from a
Perceptron freom one layer to the next layer represents a different output.
The Multi Layer Perceptron has another common name called neural network.
It basically consists of two phases Forward Propagation and Backward Propagation
Forward Propagation : It adds all the inputs multiplying by its respective wights and
generates output.
Backward Propagation: It sends the errors backward by assigning them to each
unit according to its error.If any error occurs it shows it
in three units.
1. Error in Input Unit
2. Error in Output Unit
3. Error in Hidden Unit.
In real world Perceptrons work under the framework of Deep Learning such as
PyTorch, TensorFlow. These Framework work on the Complete Philosophy
of Single Layer Perceptron and Multiple layer Perceptron and thereby constructs
the Network of Perceptrons automatically.
No comments:
Post a Comment