The main idea of this project was to develop an AI similar to MarI/O, which has the purpose of mastering the game “Super Mario World” completely on its own – by interacting and learning. At first, I thought it would be sufficient if I managed to apply the AI to other emulated games.
I want to use this blog to glance over what was accomplished so far and what the next steps will be.
So far, I did research on reinforcement learning, neural networks, evolutionary computation, the NEAT concept and the scripting language Lua. Although I believe my research to be somewhat decent, I totally see that it is very technical – and I might have done a bad job of getting a good common ground. I will leave these blogs as they are though, because it really is time to do more experiments. To this point I only took a closer look on MarI/O and its general behavior.
For the next blogs, I aim to…
- Analyze the Lua scripts for MarI/O
- Make modifications and try to observe differences in behavior
- Try to apply the scripts to other emulated games
- Try to apply the scripts to non-emulated games
- Try to apply the scripts to a website
The main idea I have with this topic is that an AI could be applied to usability tests.
