This repository contains the final project of (Fudan University) DATA130008: Introduction to Artificial Intelligence
Thanks to Dr.Zhongyu Wei, who is the Instructor of this course.
- This Project is a group work implemented by Shun Zhang, Pingxuan Huang and Donghao Li. Codes are available, if you want to utilize them, however, please indicate the source;
- If you have any question, please don't hesitate to contact 1336076538@qq.com for help.
- Some related papers are provided as references (you can obtain them at the References part);
More detailed information could be found in the Introduction.pdf;
- Students are required to implement an AI to play Gomoku;
- Gomoku Rule: Free Gomoku Rule (five or more);
- The Gomoku AIs are based on the Gomocup standard. You can create your own
AI.exethroughpisqpipe, related codes and document are here;
Please check Report.pdf for the detail information about our algorithms, AIs and experiments.
Based on 3 different algorithms, our groups successfully implemented 3 AIs:
All the algorithms can be packed into a win32 exe (less than 20M with Numpy version 1.13.1) with the help of pisqpipe. However, since the pisqpipe platform is not perfect, we provided the terminal APIs. Therefore, you can easily play with our AIs through command $python AI-name.py.
[1] Jun Hwan Kang,Hang Joon Kim, Effective Monte-Carlo Tree Search Strategies for Gomoku AI, IJCTA, 9(10), 2016, pp. 4833-4841
[2] Junru Wanga, Lan Huangb,Evolving Gomoku Solver by Genetic Algorithm,IEEE WARTIA,2014
[3] JINXING XIE and JIEFANG DONG,Heuristic Genetic Algorithms for General Capacitated Lot-Sizing Problems,2001
[4] Louis Victor Allis,Searching for Solutions in Games and Articial Intelligence,Version 8.0 of July 1, 1994