Article Instance

HTTP 200 OK
Allow: GET
Content-Type: application/json
Vary: Accept

{
    "pk": 25746,
    "title": "Tetris¬ó: Exploring Human Performance via Cross Entropy\nReinforcement Learning Models",
    "subtitle": null,
    "abstract": "What can a machine learning simulation tell us about human\nperformance in a complex, real-time task such as Tetris¬ó?\nAlthough Tetris is often used as a research tool (Mayer,\n2014), the strategies and methods used by Tetris players have\nseldom been the explicit focus of study. In Study 1, we use\ncross-entropy reinforcement learning (CERL) (Szita & Lorincz,\n2006; Thiery & Scherrer, 2009) to explore (a) the utility\nof high-level strategies (goals or objective functions) for\nmaximizing performance and (b) a variety of features and\nfeature-weights (methods) for optimizing a low-level, onezoid\noptimization strategy. Two of these optimization strategies\nquickly rise to performance plateaus, whereas two others\ncontinued towards higher but more jagged (i.e., variable)\nplateaus. In Study 2, we compare the zoid (i.e., Tetris piece)\nplacement decisions made by our best CERL models with\nthose made by the full spectrum of novice-to-expert human\nTetris players. Across 370,131 episodes collected from 67 human\nplayers, the ability of two CERL strategies to classify human\nzoid placements varied with player expertise from 43%\nfor our lowest scoring novice to around 65% for our three\nhighest scoring experts.",
    "language": "eng",
    "license": {
        "name": "",
        "short_name": "",
        "text": null,
        "url": ""
    },
    "keywords": [
        {
            "word": "Tetris"
        },
        {
            "word": "human expertise"
        },
        {
            "word": "strategies"
        },
        {
            "word": "methods"
        },
        {
            "word": "cross-entropy reinforcement learning"
        }
    ],
    "section": "Papers",
    "is_remote": true,
    "remote_url": "https://escholarship.org/uc/item/01w2w060",
    "frozenauthors": [
        {
            "first_name": "Catherine",
            "middle_name": "",
            "last_name": "Sibert",
            "name_suffix": "",
            "institution": "Rensselaer Polytechnic Institute",
            "department": ""
        },
        {
            "first_name": "Wayne",
            "middle_name": "D",
            "last_name": "Gray",
            "name_suffix": "",
            "institution": "Rensselaer Polytechnic Institute",
            "department": ""
        },
        {
            "first_name": "John",
            "middle_name": "K",
            "last_name": "Lindstedt",
            "name_suffix": "",
            "institution": "Rensselaer Polytechnic Institute",
            "department": ""
        }
    ],
    "date_submitted": null,
    "date_accepted": null,
    "date_published": "2015-01-01T10:00:00-08:00",
    "render_galley": null,
    "galleys": [
        {
            "label": "PDF",
            "type": "pdf",
            "path": "https://journalpub.escholarship.org/cognitivesciencesociety/article/25746/galley/15370/download/"
        }
    ]
}