A reinforcement learning algorithm to train a tetris playing agent

Abstract

In this paper we investigate reinforcement learning approaches for the popular computer game Tetris. User-defined reward functions have been applied to T D(0) learning based on ε-greedy strategies in the standard Tetris scenario. The numerical experiments show that reinforcement learning can significantly outperform agents utilizing fixed policies.