Options
Transfer Learning by Learning Projections from Target to Source
Publikationstyp
Conference Paper
Date Issued
2020-04
Sprache
English
Volume
12080 LNCS
Start Page
119
End Page
131
Citation
18th International Conference on Intelligent Data Analysis (IDA 2020)
Contribution to Conference
Publisher DOI
Scopus ID
Using transfer learning to help in solving a new classification task where labeled data is scarce is becoming popular. Numerous experiments with deep neural networks, where the representation learned on a source task is transferred to learn a target neural network, have shown the benefits of the approach. This paper, similarly, deals with hypothesis transfer learning. However, it presents a new approach where, instead of transferring a representation, the source hypothesis is kept and this is a translation from the target domain to the source domain that is learned. In a way, a change of representation is learned. We show how this method performs very well on a classification of time series task where the space of time series is changed between source and target.
Subjects
Boosting
Transfer learning