HyperNTM: Evolving Scalable Neural Turing Machines Through HyperNEAT

Jakob Merrild, Mikkel Angaju Rasmussen, Sebastian Risi

    Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

    Abstract

    Recent developments in memory-augmented neural networks allowed sequential problems requiring long-term memory to be solved, which were intractable for traditional neural networks. However, current approaches still struggle to scale to large memory sizes and sequence lengths. In this paper we show how access to an external memory component can be encoded geometrically through a novel HyperNEAT-based Neural Turing Machine (HyperNTM). The indirect HyperNEAT encoding allows for training on small memory vectors in a bit vector copy task and then applying the knowledge gained from such training to speed up training on larger size memory vectors. Additionally, we demonstrate that in some instances, networks trained to copy nine bit vectors can be scaled to sizes of 1,000 without further training. While the task in this paper is simple, the HyperNTM approach could now allow memory-augmented neural networks to scale to problems requiring large memory vectors and sequence lengths.
    Original languageEnglish
    Title of host publicationInternational Conference on the Applications of Evolutionary Computation : EvoApplications 2018
    Number of pages17
    PublisherSpringer
    Publication date2018
    Pages750-766
    ISBN (Electronic)978-3-319-77538-8
    DOIs
    Publication statusPublished - 2018
    SeriesLecture Notes in Computer Science
    Volume10784
    ISSN0302-9743

    Keywords

    • Memory-augmented neural networks
    • HyperNEAT
    • Neural Turing Machine
    • Scalability
    • Memory vectors

    Fingerprint

    Dive into the research topics of 'HyperNTM: Evolving Scalable Neural Turing Machines Through HyperNEAT'. Together they form a unique fingerprint.

    Cite this