Deakin University
Browse

File(s) under permanent embargo

Learning Deep Matrix Representations

journal contribution
posted on 2023-10-25, 05:30 authored by Kien Do, Truyen TranTruyen Tran, Svetha VenkateshSvetha Venkatesh
We present a new distributed representation in deep neural nets wherein the information is represented in native form as a matrix. This differs from current neural architectures that rely on vector representations. We consider matrices as central to the architecture and they compose the input, hidden and output layers. The model representation is more compact and elegant -- the number of parameters grows only with the largest dimension of the incoming layer rather than the number of hidden units. We derive several new deep networks: (i) feed-forward nets that map an input matrix into an output matrix, (ii) recurrent nets which map a sequence of input matrices into a sequence of output matrices. We also reinterpret existing models for (iii) memory-augmented networks and (iv) graphs using matrix notations. For graphs we demonstrate how the new notations lead to simple but effective extensions with multiple attentions. Extensive experiments on handwritten digits recognition, face reconstruction, sequence to sequence learning, EEG classification, and graph-based node classification demonstrate the efficacy and compactness of the matrix architectures.

History

Journal

arXiv

Article number

1703.01454

Publication classification

CN Other journal article

Publisher

Cornell University

Usage metrics

    Research Publications

    Categories

    No categories selected

    Keywords

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC