Auto-associative and hetroassociative memory

Introduction Associative memory

•  Pattern association involves associating a new pattern with a stored pattern.
  It is a “simplified” model of human memory.
•  Types of associative memory:

1.  Heteroassociative memory
2. Autoassociative memory
3. Hopfield Net
4. Bidirectional Associative Memory (BAM)

• These are usually single-layer networks.
• The neural network is firstly trained to store a set of patterns in the form s : t
  s represents the input vector and t the corresponding output vector.
• The neural network is then tested on a set of data to test its “memory” by using it to
  identify patterns containing incorrect or missing information.
• Associative memory can be feed forward or recurrent.
• Autoassociative memory cannot hold an infinite number of patterns.
• Factors that affect this:  Complexity of each pattern, Similarity of input patterns

Auto Associative Memory Architecture

Auto Associative Architecture

Auto associative Memory

• The inputs and output vectors s and t are the same.
• The Hebb rule is used as a learning  algorithm or calculate the weight matrix by summing the outer products of each input-output pair.
• The autoassociative application algorithm is used to test the algorithm

Hetero associative Memory

Hetero Associative Architecture

Hetero associative Memory

• The inputs and output vectors s and t are different.
• The Hebb rule is used as a learning algorithm or calculate the weight matrix by summing the outer products of each input-output pair.
• The heteroassociative application algorithm is used to test the algorithm.

The Hebb Algorithm

• Initialize weights to zero, wij =0, where i = 1, …, n and j = 1, …, m.
• For each training case s:t repeat:
  • xi = si , where i=1,…,n
  • yi = tj, where j = 1, .., m
  • Adjust weights wij(new) = wij(old) + xiyj,                                where i = 1, .., n and j = 1, .., m