Project Abstract

Pokemon is a popular competitive game with many unknown and random elements that could prove challenging for AI. As various types of AI tackle this problem in different ways this work seeks to compare different attempts at AI for Pokemon. An AI that implements Alpha-Beta Pruning was developed for this work and heuristic evaluations to be used by the AI were compared. Also compared were the improvements gained from using a Transposition Table and Move Ordering and the influence of search depth on performance. Finally, the AI is compared to previous Machine Learning attempts before concluding that while the AI has surpassed older models that used Q-Learning it falls short of a newer Actor-Critic model.

Keywords: Artificial Intelligence, Video Games,

 Conference Details

Session: Poster Session A at Poster Stand 115

Location: Sir Stanley Clarke Auditorium at Tuesday 7th 13:30 – 17:00

Markers: Eike Neumann, Jiaxiang Zhang

Course: BSc Computer Science, 3rd Year

Future Plans: I’m looking for work