TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Improving efficiency in convolutional neural networks with multilinear filters

Tutkimustuotosvertaisarvioitu

Yksityiskohdat

AlkuperäiskieliEnglanti
Sivut328-339
Sivumäärä12
JulkaisuNeural Networks
Vuosikerta105
DOI - pysyväislinkit
TilaJulkaistu - 1 syyskuuta 2018
OKM-julkaisutyyppiA1 Alkuperäisartikkeli

Tiivistelmä

The excellent performance of deep neural networks has enabled us to solve several automatization problems, opening an era of autonomous devices. However, current deep net architectures are heavy with millions of parameters and require billions of floating point operations. Several works have been developed to compress a pre-trained deep network to reduce memory footprint and, possibly, computation. Instead of compressing a pre-trained network, in this work, we propose a generic neural network layer structure employing multilinear projection as the primary feature extractor. The proposed architecture requires several times less memory as compared to the traditional Convolutional Neural Networks (CNN), while inherits the similar design principles of a CNN. In addition, the proposed architecture is equipped with two computation schemes that enable computation reduction or scalability. Experimental results show the effectiveness of our compact projection that outperforms traditional CNN, while requiring far fewer parameters.

Tutkimusalat

Julkaisufoorumi-taso