Genetic Programming Applications

This thesis deals with the problem of the Langton's artificial ant task and the use of tree-based and linear genetic programming as methods for solving this problem. Langton's ant is an abstract mathematical model that uses simple rules to move an ant along a two-dimensional grid. The task is to collect all the food in as few steps as possible. The goal of the thesis is the application of the two techniques mentioned above to selected instances of this problem, subsequent experimentation and statistical evaluation, including a comparison of the results with those available in the literature. In addition, the aim will be to optimize the given methods with various improvements to achieve the best possible results