Scribd
Upload
20 views

1 Neural Network: AND Function: Threshold (Y) 2

This document discusses the basics of neural networks including: 1) Examples of simple neural networks that perform logic functions like AND, OR, XOR. 2) How individual neurons are modeled with inputs, weights, activation functions and thresholds. 3) How simple networks are constructed and trained using techniques like backpropagation to adjust the weights between neurons. 4) Some key considerations in designing neural networks like network architecture, connections between layers, and training parameters. 5) Common applications of neural networks in areas like prediction and classification.

Uploaded by

damag3
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPS, PDF, TXT or read online on Scribd
20 views

1 Neural Network: AND Function: Threshold (Y) 2

This document discusses the basics of neural networks including: 1) Examples of simple neural networks that perform logic functions like AND, OR, XOR. 2) How individual neurons are modeled with inputs, weights, activation functions and thresholds. 3) How simple networks are constructed and trained using techniques like backpropagation to adjust the weights between neurons. 4) Some key considerations in designing neural networks like network architecture, connections between layers, and training parameters. 5) Common applications of neural networks in areas like prediction and classification.

Uploaded by

damag3
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPS, PDF, TXT or read online on Scribd
You are on page 1/ 13

1st Neural Network: AND function

X1 1

Y
X2 1

Threshold(Y) = 2
Y
1

2
1st Neural Network: OR function
X1 2
Y
X2 2

Threshold(Y) = 2
Y
1

2
1st Neural Network: AND not function
X1 2

Y
X2 -1

Threshold(Y) = 2
Y
1

2
1st Neural Network: XOR function
2
X1 Z1 2
-1
-1 Y
X2 Z2 2
2

Threshold(Y) = 2
Y
1

2
Modeling a neuron

 aj :Activation value of unit j


 wj,I :Weight on the link from unit j to unit i
 inI :Weighted sum of inputs to unit i
 aI :Activation value of unit i ini   j Wj , iaj

Activation Function

 Stept(x) = 1 if x >= t, else 0


 Sign(x) = +1 if x >= 0, else –1
 Sigmoid(x) = 1/(1+e-x)
Simple Network
Learning (training)
Learning in ANN is
adjusting the connection
weight between neurons.
Knowledge is stored in this
weight matrices.
Training Methods
 Unsupervised Learning (learning without
teacher or learning by doing)
 Supervised Learning (learning with
teacher)
 Back propagation
It’s similar to Supervised Learning but the error
signal is used to adjust the weight.
Simple Training
While (epoch produces an error) {
// give the next input
Error = Target – Output;
If (Error <> 0)
Weightj+1 = Weightj + LearningRate * Inputj * Error;
}

Epoch : Presentation of the entire training set to the


neural network.
Example

AND
Input X 0 0 1 1
Input Y 0 1 0 1
Output 0 0 0 1
ANN Design:
 How to arrange neurons in various layers
 How to connect neuron from different layer
 How the neurons get input(s) and give
result(s)
 Learning rate value
 Using bias or not
 How to train the ANN
Applications:
 Prediction
 Currency, stock exchange, weather, etc.
 Classification
 Pattern, shape, etc.

You might also like