Personas versus clones for player decision modeling

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

    Research output: Contribution to journalArticle

    Abstract

    The current paper investigates how to model human play styles. Building on decision and persona theory we evolve game playing agents representing human decision making styles. Two methods are developed, applied, and compared: procedural personas, based on utilities designed with expert knowledge, and clones, trained to reproduce play traces. Additionally, two metrics for comparing agent and human decision making styles are proposed and compared. Results indicate that personas evolved from designer intuitions can capture human decision making styles equally well as clones evolved from human play traces.

    Original languageEnglish (US)
    Pages (from-to)159-166
    Number of pages8
    JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume8770
    StatePublished - 2014

    Fingerprint

    Clone
    Decision making
    Modeling
    Decision Making
    Trace
    Human
    Game
    Metric
    Style

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Personas versus clones for player decision modeling. / Holmgård, Christoffer; Liapis, Antonios; Togelius, Julian; Yannakakis, Georgios N.

    In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 8770, 2014, p. 159-166.

    Research output: Contribution to journalArticle

    Holmgård, Christoffer ; Liapis, Antonios ; Togelius, Julian ; Yannakakis, Georgios N. / Personas versus clones for player decision modeling. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2014 ; Vol. 8770. pp. 159-166.
    @article{8fac7aaa8ea04cd69188255effe768a7,
    title = "Personas versus clones for player decision modeling",
    abstract = "The current paper investigates how to model human play styles. Building on decision and persona theory we evolve game playing agents representing human decision making styles. Two methods are developed, applied, and compared: procedural personas, based on utilities designed with expert knowledge, and clones, trained to reproduce play traces. Additionally, two metrics for comparing agent and human decision making styles are proposed and compared. Results indicate that personas evolved from designer intuitions can capture human decision making styles equally well as clones evolved from human play traces.",
    author = "Christoffer Holmg{\aa}rd and Antonios Liapis and Julian Togelius and Yannakakis, {Georgios N.}",
    year = "2014",
    language = "English (US)",
    volume = "8770",
    pages = "159--166",
    journal = "Lecture Notes in Computer Science",
    issn = "0302-9743",
    publisher = "Springer Verlag",

    }

    TY - JOUR

    T1 - Personas versus clones for player decision modeling

    AU - Holmgård, Christoffer

    AU - Liapis, Antonios

    AU - Togelius, Julian

    AU - Yannakakis, Georgios N.

    PY - 2014

    Y1 - 2014

    N2 - The current paper investigates how to model human play styles. Building on decision and persona theory we evolve game playing agents representing human decision making styles. Two methods are developed, applied, and compared: procedural personas, based on utilities designed with expert knowledge, and clones, trained to reproduce play traces. Additionally, two metrics for comparing agent and human decision making styles are proposed and compared. Results indicate that personas evolved from designer intuitions can capture human decision making styles equally well as clones evolved from human play traces.

    AB - The current paper investigates how to model human play styles. Building on decision and persona theory we evolve game playing agents representing human decision making styles. Two methods are developed, applied, and compared: procedural personas, based on utilities designed with expert knowledge, and clones, trained to reproduce play traces. Additionally, two metrics for comparing agent and human decision making styles are proposed and compared. Results indicate that personas evolved from designer intuitions can capture human decision making styles equally well as clones evolved from human play traces.

    UR - http://www.scopus.com/inward/record.url?scp=84921501208&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84921501208&partnerID=8YFLogxK

    M3 - Article

    AN - SCOPUS:84921501208

    VL - 8770

    SP - 159

    EP - 166

    JO - Lecture Notes in Computer Science

    JF - Lecture Notes in Computer Science

    SN - 0302-9743

    ER -