Group Project -- Gomoku
Overview
Gomoku, also called Five in a Row, is an abstract strategy board game. It
is traditionally played with Go pieces (black and white stones) on a 15×15 Go
board while in the past a 19×19 board was standard. Because pieces are
typically not moved or removed from the board, gomoku may also be played
as a paper-and-pencil game. The game is known in several countries under
different names.
Rules
Players consist of you and a robot. Players alternate turns placing a stone
of their color on an empty intersection. Color selection is random and black
plays first. The winner is the first player to form an unbroken line of five
stones of their color horizontally, vertically, or diagonally. In our rules, this
line must be exactly five stones long; six or more stones in a row does not
count as a win and is called an overline. If the board is completely filled and
no one can make a line of 5 stones, then the game ends in a draw.
Environment
Python 3.7 or above, Numpy, Pillow and other packages
you need.
Using pip install numpy and pip install Pillow to install
them.
If you don't have a GPU for training, you can use free GPUs from
l Google Colab(https://colab.research.google.com),
l Kaggle(https://www.kaggle.com/), with tutorial
https://zhuanlan.zhihu.com/p/563777405
Requirement
The UI and operation logic have been given in Gomoku.zip. Make a
group of 1~3 students and you need to
l Implement the robot(chessboard, robot_color, last_drop)
in algorithm.py and return the location of piece dropped in tuple
(r,c).
l Use alpha-beta pruning and reinforcement learning to implement the
work above. You can use Alpha-zero or build your own reinforcement
learning model.
l Compare the performance of adversarial search and reinforcement
learning, and talk about your opinion in the report.
l Please describe your results, running time, time complexity and
training process of reinforcement learning thoroughly in the report.
l Please submit a file in ZIP or RAR format containing your report
(Chinese or English) and codes on Canvas before 2023-12-31,
23:59. Name it as StudentID_Name_GroupMembers.
l Do not cheat.