projects:bildungskarenz:hexagon
This is an old revision of the document!
Table of Contents
SuperHexagon AI
- Video of the Game: https://www.youtube.com/watch?v=5mDjFdetU28
- Video of a Algorithm based Player https://www.youtube.com/watch?v=EUxRYix2aaM
Goal
Let the computer play the game.
- Vision Based, the only input it the screen image. The game is controled via keyboard emulation.
- Use a deep learning framework to train a artificial player.
- Compare different aproaches. In tearms of leaning rate and success.
- Superviced Learning
- Reinforcement Learning
- Algorithmic Aproach
- Compare the influence of preprocessing steps e.g.
- Raw image
- Polar Transformation
- Rotated (player has fixed position)
Needed Steps
- Create/Extend the current test framework.
- Automatically Start the game
- Evaluate the reuslts
- Compare different Implementations
- Robustness
- Select a sutable machine learning framework
- Collect data from human play and algorithm based implementation
- Train a Learing model
- Imporve on debug and Analytics tools.
- Iterate on the design
- Collect and analyze the Result
projects/bildungskarenz/hexagon.1597343729.txt.gz · Last modified: 2020/08/13 20:35 by wasle