ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Learning to play a real-time strategy game with deep reinforcement learning

Jernej Habjan (2019) Learning to play a real-time strategy game with deep reinforcement learning. EngD thesis.

Download (9Mb)


    With algorithm AlphaZero we have implemented the learning and recommendation of actions in a real-time strategy game. We examined a short history of deep reinforcement learning in games and summarized why the self-learning approach is best suited. For a strategic game, we determined the state of the game and transformed it with the encoder into a format suitable for learning a neural network. We determined a stopping condition with the expiry of the number of remaining moves for each player. We substantiated different configurations of learning parameters and exposed the most successful configuration for learning our game. The results were displayed with the Python Pygame module and the game engine Unreal Engine 4. In both visualizations we can play against the learned model, or we can observe two computer opponents fighting against each other.

    Item Type: Thesis (EngD thesis)
    Keywords: AlphaZero, real-time strategy game, Unreal Engine
    Number of Pages: 74
    Language of Content: Slovenian
    Mentor / Comentors:
    Name and SurnameIDFunction
    doc. dr. Matej Guid937Mentor
    prof. dr. Branko Šter283Comentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=51012&select=(ID=1538151107)
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 4373
    Date Deposited: 20 Feb 2019 15:13
    Last Modified: 13 Mar 2019 09:57
    URI: http://eprints.fri.uni-lj.si/id/eprint/4373

    Actions (login required)

    View Item