DSelect-k: Differentiable Selection in the Mixture of Experts with Applications to Multi-Task Learning (Q6369673)

From MaRDI portal





preprint article from arXiv
Language Label Description Also known as
English
DSelect-k: Differentiable Selection in the Mixture of Experts with Applications to Multi-Task Learning
preprint article from arXiv

    Statements

    7 June 2021
    0 references
    cs.LG
    0 references
    math.OC
    0 references
    stat.ML
    0 references
    Hussein Hazimeh
    0 references
    Zhe Zhao
    0 references
    Aakanksha Chowdhery
    0 references
    Maheswaran Sathiamoorthy
    0 references
    Yihua Chen
    0 references
    Rahul Mazumder
    0 references
    Lichan Hong
    0 references
    Ed H. Chi
    0 references

    Identifiers

    0 references