Further preliminary experiments with IIDLE

The immunological inspired distributed learning environment (IIDLE) is a novel learning system that belongs to the field of artificial immune systems. It is inspired by specific characteristics of the acquired immune system such as a spatially distributed population of discrete information packets (units) and internal regulatory processes. IIDLE is a relatively new platform and as such has been subject to a number of rounds exploratory experimentation to preliminary evaluate the suitability of the technique in terms of configuration and application. This work provides an additional series of preliminary and exploratory experiments along the same lines. Specifically this work investigates three potential fields of study that have been raised or rudimentarily address in previous work. They are as follows: 1. The behaviour of IIDLE whilst undergoing dynamic structural changes (adding and removing localities) (Section 2); 2. The performance of IIDLE with three embedded and competing proliferation strategies (GA PSO and ACO) (Section 3); 3. The potential for IIDLE on function approximation and classification problems where there is not a one-to-one mapping between units and solutions (Section 4). Finally Section 5 provides concluding remarks and a discussion of future research specifically ideas on how the work from the three experiments can be extended [Introduction].