ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Pruning neural network using matrix factorization

Teja Roštan (2015) Pruning neural network using matrix factorization. EngD thesis.

Download (1120Kb)


    Matrix factorization and the procedure of data fusion are used to detect patterns in data. The factorized model maps the data to a low-dimensional space, therefore shrinking it and partially eliminating noise. Factorized models are thus more robust and have a higher predictive accuracy. With this procedure we could solve the problem of overfitting in neural networks and improve their ability to generalize. Here, we report on how to simultaneously factorize the parameters of a neural network, which can be represented with multiple matrices, to prune not important connections and therefore improve predictive accuracy. We report on empirical results of pruning normal and deep neural networks. The proposed method performs similarly to the best standard approaches to pruning neural networks.

    Item Type: Thesis (EngD thesis)
    Keywords: neural networks, matrix factorization, pruning
    Number of Pages: 52
    Language of Content: Slovenian
    Mentor / Comentors:
    Name and SurnameIDFunction
    doc. dr. Tomaž Curk299Mentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=51012&select=(ID=1536482243 )
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 3061
    Date Deposited: 09 Sep 2015 15:53
    Last Modified: 18 Sep 2015 09:48
    URI: http://eprints.fri.uni-lj.si/id/eprint/3061

    Actions (login required)

    View Item