A New Schema and Landscape for Programs

Dr Kaur and I sent a paper on a new schema and landscape for IEEE TEC peer review two weeks ago.

In the paper we showcase the new schema and landscape analyses by applying it to the Santa Fe ant problem. This caused us to discover for the first time the relationship between program structures and program fitness. Traditionally the Santa Fe ant problem is well known for presenting a random fitness relationship when analyzed by any other method.

We also show for the first time the systematic approaches to fitness improvement that programs make during genetic programming runs, thereby showing that the process is very different from what a random search does. We test a new variation and representation method that were designed based on our findings and obtained more efficient evolutionary search.

Please send me an email if you would like a copy for personal review.

We are currently undecided on whether we should post the pre-review paper to Arxiv.org. Any advice?