Learning Tetris with reinforcement learning
Year: 2021
Type: dissertation
Abstract: The last decades, scientists have expressed an increasing interest for the game Tetris and \nmore precisely for efficient algorithms that can score the most in-game points. \nA plethora of approaches have been tried out, including genetic algorithms, linear programming, cross-entropy and natural policy gradient, but none of them competes with \nexperts players playing under no time pressure. \nIn recent years, scientists have started applying reinforcement learning in Tetris as it \ndisplays effective results in adapting to video game environments, exploit mechanisms \nand deliver extreme performances. \nCurrent thesis aims to introduce Memory Based Learning, a reinforcement learning algorithm which uses a memory that helps in the training process by replaying past experiences.
Source: Dione (University of Piraeus)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3214945965