Mixed-Initiative Level Design with RL Brush

Omar Delarosa, Hang Dong, Mindy Ruan, Ahmed Khalifa, Julian Togelius

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    This paper introduces RL Brush, a level-editing tool for tile-based games designed for mixed-initiative co-creation. The tool uses reinforcement-learning-based models to augment manual human level-design through the addition of AI-generated suggestions. Here, we apply RL Brush to designing levels for the classic puzzle game Sokoban. We put the tool online and tested it in 39 different sessions. The results show that users using the AI suggestions stay around longer and their created levels on average are more playable and more complex than without.

    Original languageEnglish (US)
    Title of host publicationArtificial Intelligence in Music, Sound, Art and Design - 10th International Conference, EvoMUSART 2021, Held as Part of EvoStar 2021, Proceedings
    EditorsJuan Romero, Tiago Martins, Nereida Rodríguez-Fernández
    PublisherSpringer Science and Business Media Deutschland GmbH
    Pages412-426
    Number of pages15
    ISBN (Print)9783030729134
    DOIs
    StatePublished - 2021
    Event10th International Conference on Artificial Intelligence in Music, Sound, Art and Design, EvoMUSART 2021 held as Part of EvoStar 2021 - Virtual, Online
    Duration: Apr 7 2021Apr 9 2021

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume12693 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference10th International Conference on Artificial Intelligence in Music, Sound, Art and Design, EvoMUSART 2021 held as Part of EvoStar 2021
    CityVirtual, Online
    Period4/7/214/9/21

    Keywords

    • Mixed initiative tools
    • Procedural content generation
    • Reinforcement learning

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • General Computer Science

    Fingerprint

    Dive into the research topics of 'Mixed-Initiative Level Design with RL Brush'. Together they form a unique fingerprint.

    Cite this