Emergent Thought Automated Agent Design Mini-project Report

The following report presents the preliminary results and theoretical backgrounds of a multi-layer evolutionary mechanism. The project aims at developing a mechanism for automated agent design within a pre-designed multi-layer architecture. The core mechanism used in the experiments is a Genetic Programming algorithm working in a specially designed condition-action rules system. There is described an extension to Genetic Programming, which is based on a multi-layer evolution paradigm, inspired by notion of the niche & design spaces. The multi-layer GP algorithm is distributed across a number of workstations and works in parallel. There are presented plans for experiments with the multi-layer co-evolutionary mechanism for parallel distributed search within restrained search space. Attention is paid to evolutionary mechanisms in general, with no commitments to any particular problem. In theory the multi-layer evolutionary process can be applied to a number of systems and problems, e.g. to Genetic Programming. The experiments are based on simple Artiicial Life scenario in which agents act in a continuous 3D environment. The simulation is 2D online and 3D oo-line (as a movie). The use of agents in entertainment as a perfect environment for agent research is stressed.