Thesis Details

Aproximace hlubokých neuronových sítí

Bachelor's Thesis Student: Stodůlka Martin Academic Year: 2018/2019 Supervisor: Vaverka Filip, Ing.
English title
Deep Neural Networks Approximation
Language
Czech
Abstract

The goal of this work is to find out the impact of approximated computing on accuracy of deep neural network, specifically neural networks for image classification. A version of framework Caffe called Ristretto-caffe was chosen for neural network implementation, which was extended for the use of approximated operations. Approximated computing was used for multiplication in forward pass for convolution. Approximated components from Evoapproxlib were chosen for this work.

Keywords

Deep neural networks, image classification, approximated computing, approximated circuits, Evoapproxlib, C++, CUDA, Caffe, Ristretto-caffe

Department
Department of Computer Systems FIT BUT
Degree Programme
Information Technology
Files
Status
defended, grade D
Date
13 June 2019
Reviewer
Mrázek Vojtěch, Ing., Ph.D.
Committee
Zbořil František, doc. Ing., Ph.D. (DITS FIT BUT), předseda
Burget Lukáš, doc. Ing., Ph.D. (DCGM FIT BUT), člen
Grézl František, Ing., Ph.D. (DCGM FIT BUT), člen
Hliněná Dana, doc. RNDr., Ph.D. (DMAT FEEC BUT), člen
Strnadel Josef, Ing., Ph.D. (DCSY FIT BUT), člen
Citation
STODŮLKA, Martin. Aproximace hlubokých neuronových sítí. Brno, 2019. Bachelor's Thesis. Brno University of Technology, Faculty of Information Technology. 2019-06-13. Supervised by Vaverka Filip. Available from: https://www.fit.vut.cz/study/thesis/21819/
BibTeX 
@bachelorsthesis{FITBT21819,
    author = "Martin Stod\r{u}lka",
    type = "Bachelor's thesis",
    title = "Aproximace hlubok\'{y}ch neuronov\'{y}ch s\'{i}t\'{i}",
    school = "Brno University of Technology, Faculty of Information Technology",
    year = 2019,
    location = "Brno, CZ",
    language = "czech",
    url = "https://www.fit.vut.cz/study/thesis/21819/"
}

Theses

Back to top