Abstract
dc:descriptionIn this paper we examine the relevance of transfer learning in deep learning context, we review different techniques that are used when executing and improving transfer learning tasks and we review the uses of transfer learning in cases such as computer vision, where learning requires a lot of data, computing resources and time and therefore this problem can be solved with transfer learning. After examining the uses and methods of transfer learning, experiments that were done in other research papers as well as experiments done for this paper, a conclusion became apparent that transfer learning can decrease the amount of data required, speed up the training process as well as increase the accuracy of neural networks.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Maziukas, Indiras,
- Contributors dc:contributor
-
- Dapkūnas, Boleslovas
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD81706480&prefLang=en_US