Information Technology Reference
In-Depth Information
5.3 MOP-Ideal Test Bed for Immune Mechanism Simulation
Figure 2 (b) is a reflection of the immune response from Ab when stimulated by Ag.
It can be seen that the population (Ab concentration) keeps increasing with the pres-
ence of antigenic stimulus until a stable concentration level is achieved. If without
local extrema, then a problem (i.e. ZDT1~ZDT3) can be regarded as an unvaccinated
immune system (whose Ab concentration bears characteristics illustrated in the first
three graphs in Figure 2 (b), and such a characteristic is seen as primary immune
response). On the other hand, when a problem has many local extrema and these ex-
trema share some resemblances (ZDT4), it corresponds to an immune system with
continuous vaccinations. As in the last graph of Figure 2 (b), the Ab concentration
initially reacts as a primary response, however, in the following vaccinations the peak
values match each set of extrema and this is recognized here as secondary response.
Therefore, if a test problem is adequately designed according to the above principle,
MOP will be an ideal test bed for the immune mechanism simulations.
6 Conclusions and Further Research
Our conclusion is that, as a solution to a MOP, AIS offers advantages over traditional
population-based GA schemes. Such superiority is based on the fact that AIS is in-
spired by a different regime of natural mechanisms. As a result, one could identify
two directions for future research; one is to improve PAIA such as its mutation opera-
tor and termination condition. The other is to further compare and understand the
differences between AIS and GA so that one can be confident in deciding which one
is more suitable to handle a specific problem.
Acknowledgements
The authors would like to thank Dr. Fabio Freschi for his kind help in providing the
results of applying VIS to the ZDT1~ZDT4 test suite.
References
1.
Ishida, Y.: Fully Distributed Diagnosis by PDP Learning Algorithm: Towards Immune
Network PDP Model. Proc. of the IEEE International Joint Conference on Neural Net-
works. San Diego, USA (1990) 777-782
2.
Forrest, S., Perelson, A. S., Allen, L., Cherukuri, R.: Self-Nonself Discrimination in a
Computer. Proc. Of IEEE Symposium on Research in Security and Privacy. Oakland, USA
(1994) 202-212
3.
de Castro, L. N., Von Zuben, F. J.: aiNet: An Artificial Immune Network for Data Analy-
sis In: Abbass, H. A., Sarker, R. A., Newton, C. S. (eds.): Data Mining: A Heuristic Ap-
proach. Idea Group Publishing, USA (2001) 231-259
4.
Timmis, J.: Artificial Immune Systems: A Novel Data Analysis Technique Inspired by the
Immune Network Theory. PhD Dissertation, Department of Computer Science, University
of Wales (2000)
