Summary
This is a project done as part of my bachelor's thesis on Computer Science. The idea was to make a simple fighting game prototype and develop an AI that can learn how to play with the player through Deep Reinforcement Learning.
As it stands now the prototype consists of a really basic fighting game with a light and heavy attack, and a block button similar to Mortal Kombat. It can be played as Player vs RL-AI or it can be done as RL-AI vs regular AI.
Responsibilities
- Developing the fighting game mechanics
- Determining AI parameters and network config
- Adapting the learning algorithm
- Bug fixing
Technologies Used
- Unity Engine
- Git
- Python
- Python.NET
