The motivation for this research is to explore how agents perform in real-time strategic game environments and whether they can easily surpass human capabilities. Additionally, we aim to move beyond the traditional Atari game framework to demonstrate that reinforcement learning can be applied more broadly, not limited to just Atari games. We have chosen Tetris, a game with simpler and more understandable rules and not an Atari game, as our environment. This allows us to focus not on complex game mechanics but on the application aspect.
https://github.com/Felixqaq/Implementing-Tetris-Game-with-Reinforcement-Learning