Zpět na vyhledávání

Mgr. Karol Kuna

Bachelor's thesis

Predikce časových řad pomocí neuronových sítí

Time Series Prediction Using Neural Networks
Abstract:
Táto práca porovnáva existujúce metódy predikcie časových radov v reálnom čase pomocou neurónových sietí. Dôraz sa kladie na rekurentné neurónové siete a algoritmy učiace sa online, ako napr. real-time recurrent learning a truncated backpropagation through time. Okrem štandardnej Elmanovej rekurentnej siete je preskúmaná Clockwork rekurentná sieť. Metódy sú porovnané vzhľadom na presnosť predpovedí …more
Abstract:
This thesis compares existing methods for predicting time series in real time using neural networks. Focus is put on recurrent neural networks (RNNs) and online learning algorithms, such as Real-Time Recurrent Learning and truncated Backpropagation Through Time. In addition to the standard Elman’s RNN architecture, Clockwork-RNN is examined. Methods are compared in terms of prediction accuracy and …more
 

Keywords

time series prediction neural network recurrent network backpropagation backpropagation through time real-time recurrent learning clockwork recurrent network časové rady predikcia neurónové siete rekurentné siete
 
Language used: English
Date on which the thesis was submitted / produced: 18. 5. 2015
Identifier: https://is.muni.cz/th/x3epf/

Thesis defence

  • Date of defence: 16. 6. 2015
  • Supervisor: doc. RNDr. Tomáš Brázdil, Ph.D.
  • Reader: RNDr. Karel Vaculík

Citation record

Cite this text

Full text of thesis

Contents of on-line thesis archive
Published in Theses:
  • světu
Other ways of accessing the text
Institution archiving the thesis and making it accessible: Masarykova univerzita, Fakulta informatiky
Složka Reference to the local database directory of the institution