Development of a benchmarking framework for Inverse Reinforcement Learning algorithms based on Tetris
Pascal Bock · OPUS Publication Server of the University of Stuttgart (University of Stuttgart) · 2015
Tetris is one of the oldest, most popular and most well-known video games. The simple rules and scoring options make it a viable choice for benchmarking artificial intelligence, especially in the machine learning department. This work describes a customizable benchmarking framework using a simplified variant of the original Tetris game focused on Inverse Reinforcement Learning algorithms.