Preliminary experiments with IIDLE

The mammalian acquired immune system is a robust and powerful information processing system that demonstrates features such as decentralised control, parallel processing adaptation and learning through experience. Artificial Immune Systems (AIS) are a class of machine-learning algorithms that are imbued with some principles of the immune system and attempt to take advantage of some of the benefits of the features of the immune system to tackle difficult problem domains. A novel artificial immune system called IIDLE -- the immunological inspired distributed learning environment -- has been introduced previously in regard to the techniques inspiration conceptualisation and rudimentary architecture and processes. IIDLE is inspired by the spatially distributed nature of the acquired immune system and the clonal selection principle that describes how the immune system learns and adapts in response to stimulation from its environment. In this work the IIDLE platform is preliminarily put to the test on simple combinatorial optimisation (TSP) and function optimisation problems (Schwefel's function). A number of experiments are performed to gain an initial understanding of the suitability and applicability of IIDLE in the context of dynamic function optimisation multiple constraints complementing objective functions and human interactive search. Some interesting and other surprising results are achieved that perhaps provide a first indication of the potential of this novel machine learning system. Further work is required and a number of exciting additional experiments are proposed.