Procedural personas as critics for dungeon generation

Antonios Liapis, Christoffer Holmgård, Georgios N. Yannakakis, Julian Togelius

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

    Abstract

    This paper introduces a constrained optimization method which uses procedural personas to evaluate the playability and quality of evolved dungeon levels. Procedural personas represent archetypical player behaviors, and their controllers have been evolved to maximize a specific utility which drives their decisions. A “baseline” persona evaluates whether a level is playable by testing if it can survive in a worst-case scenario of the playthrough. On the other hand, a Monster Killer persona or a Treasure Collector persona evaluates playable levels based on how many monsters it can kill or how many treasures it can collect, respectively. Results show that the implemented two-population genetic algorithm discovers playable levels quickly and reliably, while the different personas affect the layout, difficulty level and tactical depth of the generated dungeons.

    Original languageEnglish (US)
    Title of host publicationApplications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings
    PublisherSpringer Verlag
    Pages331-343
    Number of pages13
    Volume9028
    ISBN (Print)9783319165486
    DOIs
    StatePublished - 2015
    Event18th European Conference on the Applications of Evolutionary Computation, EvoApplications 2015 - Copenhagen, Denmark
    Duration: Apr 8 2015Apr 10 2015

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume9028
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other18th European Conference on the Applications of Evolutionary Computation, EvoApplications 2015
    CountryDenmark
    CityCopenhagen
    Period4/8/154/10/15

    Fingerprint

    Constrained optimization
    Genetic algorithms
    Controllers
    Evaluate
    Testing
    Population Genetics
    Constrained Optimization
    Optimization Methods
    Layout
    Baseline
    Maximise
    Genetic Algorithm
    Controller
    Scenarios

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Liapis, A., Holmgård, C., Yannakakis, G. N., & Togelius, J. (2015). Procedural personas as critics for dungeon generation. In Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings (Vol. 9028, pp. 331-343). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9028). Springer Verlag. https://doi.org/10.1007/978-3-319-16549-3_27

    Procedural personas as critics for dungeon generation. / Liapis, Antonios; Holmgård, Christoffer; Yannakakis, Georgios N.; Togelius, Julian.

    Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings. Vol. 9028 Springer Verlag, 2015. p. 331-343 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9028).

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

    Liapis, A, Holmgård, C, Yannakakis, GN & Togelius, J 2015, Procedural personas as critics for dungeon generation. in Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings. vol. 9028, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9028, Springer Verlag, pp. 331-343, 18th European Conference on the Applications of Evolutionary Computation, EvoApplications 2015, Copenhagen, Denmark, 4/8/15. https://doi.org/10.1007/978-3-319-16549-3_27
    Liapis A, Holmgård C, Yannakakis GN, Togelius J. Procedural personas as critics for dungeon generation. In Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings. Vol. 9028. Springer Verlag. 2015. p. 331-343. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-16549-3_27
    Liapis, Antonios ; Holmgård, Christoffer ; Yannakakis, Georgios N. ; Togelius, Julian. / Procedural personas as critics for dungeon generation. Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings. Vol. 9028 Springer Verlag, 2015. pp. 331-343 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{4bd9a93ee17849288a005270731fa399,
    title = "Procedural personas as critics for dungeon generation",
    abstract = "This paper introduces a constrained optimization method which uses procedural personas to evaluate the playability and quality of evolved dungeon levels. Procedural personas represent archetypical player behaviors, and their controllers have been evolved to maximize a specific utility which drives their decisions. A “baseline” persona evaluates whether a level is playable by testing if it can survive in a worst-case scenario of the playthrough. On the other hand, a Monster Killer persona or a Treasure Collector persona evaluates playable levels based on how many monsters it can kill or how many treasures it can collect, respectively. Results show that the implemented two-population genetic algorithm discovers playable levels quickly and reliably, while the different personas affect the layout, difficulty level and tactical depth of the generated dungeons.",
    author = "Antonios Liapis and Christoffer Holmg{\aa}rd and Yannakakis, {Georgios N.} and Julian Togelius",
    year = "2015",
    doi = "10.1007/978-3-319-16549-3_27",
    language = "English (US)",
    isbn = "9783319165486",
    volume = "9028",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "331--343",
    booktitle = "Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings",

    }

    TY - GEN

    T1 - Procedural personas as critics for dungeon generation

    AU - Liapis, Antonios

    AU - Holmgård, Christoffer

    AU - Yannakakis, Georgios N.

    AU - Togelius, Julian

    PY - 2015

    Y1 - 2015

    N2 - This paper introduces a constrained optimization method which uses procedural personas to evaluate the playability and quality of evolved dungeon levels. Procedural personas represent archetypical player behaviors, and their controllers have been evolved to maximize a specific utility which drives their decisions. A “baseline” persona evaluates whether a level is playable by testing if it can survive in a worst-case scenario of the playthrough. On the other hand, a Monster Killer persona or a Treasure Collector persona evaluates playable levels based on how many monsters it can kill or how many treasures it can collect, respectively. Results show that the implemented two-population genetic algorithm discovers playable levels quickly and reliably, while the different personas affect the layout, difficulty level and tactical depth of the generated dungeons.

    AB - This paper introduces a constrained optimization method which uses procedural personas to evaluate the playability and quality of evolved dungeon levels. Procedural personas represent archetypical player behaviors, and their controllers have been evolved to maximize a specific utility which drives their decisions. A “baseline” persona evaluates whether a level is playable by testing if it can survive in a worst-case scenario of the playthrough. On the other hand, a Monster Killer persona or a Treasure Collector persona evaluates playable levels based on how many monsters it can kill or how many treasures it can collect, respectively. Results show that the implemented two-population genetic algorithm discovers playable levels quickly and reliably, while the different personas affect the layout, difficulty level and tactical depth of the generated dungeons.

    UR - http://www.scopus.com/inward/record.url?scp=84925867817&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84925867817&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-16549-3_27

    DO - 10.1007/978-3-319-16549-3_27

    M3 - Conference contribution

    AN - SCOPUS:84925867817

    SN - 9783319165486

    VL - 9028

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 331

    EP - 343

    BT - Applications of Evolutionary Computation - 18th European Conference, EvoApplications 2015, Proceedings

    PB - Springer Verlag

    ER -