Project Abstract

Optimizing optimisation has been an ongoing battle since technology emerged, with scientists and mathematicians all over the world seeking more efficient and effective methods of gaining optimum results with ease. Since early 1960s when Jonn H. Holland first introduced the concept of Genetic algorithms (GA?��s) (Holland, 1992) they have presented a promising avenue in this pursuit. I set out to research the application of (GA?��s) and its ability to solve a complex resource collection and army composition problem in the context of village survival and warfare strategies.Beginning with a review of genetic algorithm literature, I establish a base understanding of the principles and methodologies used in GA?��s. I analyse the different use cases for GA?��s in both games and biology where these algorithms have exceeded beyond others. I then tailor a GA to the specific and nuance strategy in a two-phase village strategy game. Through extensive experimentation and refinement, I analyse the algorithm?��s ability when faced with different game scenarios.Central to my investigation is the use of the optimised genetic algorithm to explore the complex search space of the village resource management game. By iteratively evolving strategies using a refined process the GA aims to maximise resource assortment and optimise military strength. Through experimentation, I analyse the algorithms performance in terms of solution speed, quality, and adaptability of evolving game mechanics.My research findings offer an insight into the application of GA?��s in strategic game design and outline the importance of a correct variables when faced with a complex strategic challenge. Furthermore, I uncover the unique strategies unearthed by the GA which are not apparent to humans.My output includes the documentation of the GA design process along with the subsequent results and implications. I highlight my contributions to the field and offer recommendations to the potential future development of GA?��s in the strategic problem solving domain.

Keywords: Machine Learning, Genetic Algorithms, Strategy Games

 Conference Details

Session: Poster Session B at Poster Stand 81

Location: Sir Stanley Clarke Auditorium at Wednesday 8th 09:00 – 12:30

Markers: Manlio Valenti, Trang Doan

Course: BSc Computer Science, 3rd Year

Future Plans: I’m looking for work