ASSOCIATIVE LEARNING

CONTENT
 Definition
 History of Hebbian Learning

 Associative Memory Networks

 Types of Associative Network

 Advantage of Hebbian Learning

DEFINITION

Associative learning is a theory that

states that ideas reinforce each other
and can be linked to one another.
This lesson will explain the theory of
Associative learning as well as provide
some practical, real-life examples of
this type of learning.
Hebbian Learning
 It was proposed in 1950 by “Donald Hebb
 Hebb’s principle can be described how to alter the
weight between modal neurons
 The Weight between two neurons increase if two
neurons activate simultaneously, and reduces if
activate separately .
 If there are no signal correlation ,the weight
remain unchanged.
In case of associative network, Hebbian
learning is applied,updating the weight
Wij by Wij.

Yj
Wij
Xi
Wij = CXiYj

THE HUBB RULE

 The formulaic description of Hebbian learning,

Wij = CXiYj

Where,

Wij : is the weight from ith to jth neoron.

C : is the learning constant.
Xi : is the input signal.
yi : is the output signal.
 Associative Memory Networks

 It is the type of a neural network

 Work on basis of pattern association

 It makes a parallel search

 It is also known as Content-Addresseble Memory (CAM)

 Architecture Of Associative Networks

 In this single-layer networks are used.

 In feed forword net,information flows from input
units to output units.
 In recurrent net, connections among the nodes that
form closed loops.
 The learning rule used (Hubb Rule).

 The data and weight representation can be binary or

bipolar form.
Types Of Associative Networks

 Auto Associative Memory

 Hetero Associative Memory

 Auto-Associative Networks
 This is a single layer neural network in
which the input training vector and the
output target vector are the same.

 In this the network stores a set of

patterns.

 For this network, Hebb rule are used.

 Architecture
The architecture of auto associative network has n number
of input vectors and n number of output vectors.

Training Algorithm

 It is using the Hebb learning rule.

 Step 1 − Initialize all the weights to zero as

  wij = 0 (i = 1 to n, j = 1 to n)

 Step 2 − Perform steps 3-4 for each input vector.

 Step 3 − Activate each input unit as follows −
xi=si(i=1 to n)

 Step 4 − Activate each output unit as follows −

yj=sj(j=1 to n)

 Step 5 − Adjust the weights as follows −

wij(new)=wij(old)+xiyj
Hetero-Associative Networks
 It is also a single layer neural network
 In which the input training vector and the
output target vector are not the same.
 It is static in nature
 It would be no non-linear and delay operations.
 For this network also, Hebb rule are used
Architecture
 The architecture of Hetero Associative Memory network
has ‘n’ number of input training vectors and ‘m’ number of output
target vectors.

Training Algorithm

 It is using the Hebb or Delta learning rule.

 Step 1 − Initialize all the weights to zero as 

wij = 0 (i = 1 to n, j = 1 to m)

 Step 2 − Perform steps 3-4 for each input vector.

 Step 3 − Activate each input unit as follows −
xi = si(i = 1 to n)

 Step 4 − Activate each output unit as follows −

yj = sj(j = 1 to m)

 Step 5 − Adjust the weights as follows −

wij(new) = wij(old)+xiyj
 Advantage:-

o The main advantage of associative learning is that it is

used in the pattern association and pattern recognition
Thanks