The Mario Ai benchmark and competitions

Sergey Karakovskiy, Julian Togelius

    Research output: Contribution to journalArticle

    Abstract

    This paper describes the Mario AI benchmark, a game-based benchmark for reinforcement learning algorithms and game AI techniques developed by the authors. The benchmark is based on a public domain clone of Nintendo's classic platform game Super Mario Bros, and completely open source. During the last two years, the benchmark has been used in a number of competitions associated with international conferences, and researchers and students from around the world have contributed diverse solutions to try to beat the benchmark. The paper summarizes these contributions, gives an overview of the state of the art in Mario-playing AIs, and chronicles the development of the benchmark. This paper is intended as the definitive point of reference for those using the benchmark for research or teaching.

    Original languageEnglish (US)
    Article number6156425
    Pages (from-to)55-67
    Number of pages13
    JournalIEEE Transactions on Computational Intelligence and AI in Games
    Volume4
    Issue number1
    DOIs
    StatePublished - Mar 2012

    Fingerprint

    Reinforcement learning
    Learning algorithms
    Teaching
    Students

    Keywords

    • Benchmarking
    • competitions
    • game AI
    • reinforcement learning

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Software
    • Control and Systems Engineering
    • Electrical and Electronic Engineering

    Cite this

    The Mario Ai benchmark and competitions. / Karakovskiy, Sergey; Togelius, Julian.

    In: IEEE Transactions on Computational Intelligence and AI in Games, Vol. 4, No. 1, 6156425, 03.2012, p. 55-67.

    Research output: Contribution to journalArticle

    Karakovskiy, Sergey ; Togelius, Julian. / The Mario Ai benchmark and competitions. In: IEEE Transactions on Computational Intelligence and AI in Games. 2012 ; Vol. 4, No. 1. pp. 55-67.
    @article{f2d707c683cb42cea343bb5fb859baba,
    title = "The Mario Ai benchmark and competitions",
    abstract = "This paper describes the Mario AI benchmark, a game-based benchmark for reinforcement learning algorithms and game AI techniques developed by the authors. The benchmark is based on a public domain clone of Nintendo's classic platform game Super Mario Bros, and completely open source. During the last two years, the benchmark has been used in a number of competitions associated with international conferences, and researchers and students from around the world have contributed diverse solutions to try to beat the benchmark. The paper summarizes these contributions, gives an overview of the state of the art in Mario-playing AIs, and chronicles the development of the benchmark. This paper is intended as the definitive point of reference for those using the benchmark for research or teaching.",
    keywords = "Benchmarking, competitions, game AI, reinforcement learning",
    author = "Sergey Karakovskiy and Julian Togelius",
    year = "2012",
    month = "3",
    doi = "10.1109/TCIAIG.2012.2188528",
    language = "English (US)",
    volume = "4",
    pages = "55--67",
    journal = "IEEE Transactions on Computational Intelligence and AI in Games",
    issn = "1943-068X",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "1",

    }

    TY - JOUR

    T1 - The Mario Ai benchmark and competitions

    AU - Karakovskiy, Sergey

    AU - Togelius, Julian

    PY - 2012/3

    Y1 - 2012/3

    N2 - This paper describes the Mario AI benchmark, a game-based benchmark for reinforcement learning algorithms and game AI techniques developed by the authors. The benchmark is based on a public domain clone of Nintendo's classic platform game Super Mario Bros, and completely open source. During the last two years, the benchmark has been used in a number of competitions associated with international conferences, and researchers and students from around the world have contributed diverse solutions to try to beat the benchmark. The paper summarizes these contributions, gives an overview of the state of the art in Mario-playing AIs, and chronicles the development of the benchmark. This paper is intended as the definitive point of reference for those using the benchmark for research or teaching.

    AB - This paper describes the Mario AI benchmark, a game-based benchmark for reinforcement learning algorithms and game AI techniques developed by the authors. The benchmark is based on a public domain clone of Nintendo's classic platform game Super Mario Bros, and completely open source. During the last two years, the benchmark has been used in a number of competitions associated with international conferences, and researchers and students from around the world have contributed diverse solutions to try to beat the benchmark. The paper summarizes these contributions, gives an overview of the state of the art in Mario-playing AIs, and chronicles the development of the benchmark. This paper is intended as the definitive point of reference for those using the benchmark for research or teaching.

    KW - Benchmarking

    KW - competitions

    KW - game AI

    KW - reinforcement learning

    UR - http://www.scopus.com/inward/record.url?scp=84858967765&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84858967765&partnerID=8YFLogxK

    U2 - 10.1109/TCIAIG.2012.2188528

    DO - 10.1109/TCIAIG.2012.2188528

    M3 - Article

    AN - SCOPUS:84858967765

    VL - 4

    SP - 55

    EP - 67

    JO - IEEE Transactions on Computational Intelligence and AI in Games

    JF - IEEE Transactions on Computational Intelligence and AI in Games

    SN - 1943-068X

    IS - 1

    M1 - 6156425

    ER -