Many Faces of Representation Learning

Nada Lavrač, Vid Podpečan and Marko Robnik-Šikonja
Additional contact information
Nada Lavrač: Jožef Stefan Institute, Department of Knowledge Technologies
Vid Podpečan: Jožef Stefan Institute, Department of Knowledge Technologies
Marko Robnik-Šikonja: University of Ljubljana, Faculty of Computer and Information Science

Chapter Chapter 7 in Representation Learning, 2021, pp 153-158 from Springer

Abstract: Abstract As this monograph demonstrates, albeit propositionalization and embeddings represent two different families of data transformations, they can be viewed as the two sides of the same coin. Their main unifying element is that they transform the input data into a tabular format and express the relations between objects in the original space as distances (and directions) in the target vector space. Our work indicates that both propositionalization and embeddings address transformations of entities with defined similarity functions as a multifaceted approach to feature construction. Based on the work by Lavrač et al. (2020), this chapter explores the similarities and differences of propositionalization and embeddings in terms of data representation, learning and use, in Sects. 7.1, 7.2 and 7.3, respectively. In Sect. 7.4 we summarize the strengths and limitations of propositionalization and embeddings, and conclude this monograph with some hints for further research.

Date: 2021
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-030-68817-2_7

Ordering information: This item can be ordered from
http://www.springer.com/9783030688172

DOI: 10.1007/978-3-030-68817-2_7

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().