Controllable Path of Destruction

Matthew Siper, Sam Earle, Zehua Jiang, Ahmed Khalifa, Julian Togelius

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


    Path of Destruction (PoD) is a self-supervised method for learning iterative generators. The core idea is to produce a training set by destroying a set of artifacts, and for each destructive step create a training instance based on the corresponding repair action. A generator trained on this dataset can then generate new artifacts by "repairing"from arbitrary states. The PoD method is very data-efficient in terms of original training examples and well-suited to functional artifacts composed of categorical data, such as game levels and discrete 3D structures. In this paper, we extend the Path of Destruction method to allow designer control over aspects of the generated artifacts. Controllability is introduced by adding conditional inputs to the state-action pairs that make up the repair trajectories. We test the controllable PoD method in a 2D dungeon setting, as well as in the domain of small 3D Lego cars.

    Original languageEnglish (US)
    Title of host publicationProceedings of the 2023 IEEE Conference on Games, CoG 2023
    PublisherIEEE Computer Society
    ISBN (Electronic)9798350322774
    StatePublished - 2023
    Event5th Annual IEEE Conference on Games, CoG 2023 - Boston, United States
    Duration: Aug 21 2023Aug 24 2023

    Publication series

    NameIEEE Conference on Computatonal Intelligence and Games, CIG
    ISSN (Print)2325-4270
    ISSN (Electronic)2325-4289


    Conference5th Annual IEEE Conference on Games, CoG 2023
    Country/TerritoryUnited States


    • Controllability
    • Data Augmentation
    • Procedural Content Generation
    • Repair Function
    • Supervised Learning

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Computer Graphics and Computer-Aided Design
    • Computer Vision and Pattern Recognition
    • Human-Computer Interaction
    • Software


    Dive into the research topics of 'Controllable Path of Destruction'. Together they form a unique fingerprint.

    Cite this