1. Randomly initialise the network population;
2. For each antigenic pattern in P apply the CLONALG algorithm
that will return a set of memory cells (M*) and their coordinates
for the current antigen;
3. Determine the affinity (degree of matching) among all the
individuals of M*;
4. Eliminate all but one of the individuals in M* whose affinities
are greater than a given threshold. The purpose of this process
is to eliminate redundancy in the network by suppressing selfrecognising
elements;
5. Concatenate the remaining individuals of the previous step
with the remaining individuals found for each antigenic pattern
presented. This will result in a large population of memory
individuals M;
6. Determine the affinity of the whole population M and suppress
all but one of the self-recognising elements. This will result in
a reduced final population of memory cells that recognise and
follow the spatial distribution of the antigens.
7. Repeat Steps 2 to 6 until a pre-defined stopping criterion is met,
such as a minimum pattern recognition or classification error.