Learning Embeddings from Probabilistic Triplet Comparisons
No Thumbnail Available
URL
Journal Title
Journal ISSN
Volume Title
Perustieteiden korkeakoulu |
Master's thesis
Authors
Date
2018-10-08
Department
Major/Subject
Networking Technology
Mcode
ELEC3029
Degree programme
Master’s Programme in Computer, Communication and Information Sciences
Language
en
Pages
79+2
Series
Abstract
Learning from relative similarity comparisons has gained interest in the data science community in the past 20 years. We introduce a new way to capture relative similarity comparisons called probabilistic triplets that alleviates extreme decisions under high uncertainty, and provides finer-grained information than ordinary triplets. We describe a new method called t-SPTE that finds an embedding of objects in a Euclidean space using probabilistic triplets datasets as its input. The problem is formulated as a least squares optimization of differences between the labeled triplet probabilities and the triplet probabilities coming from the stochastic neighborhood model in the embedding space. We experimentally show that our approach improves upon previous methods, notably t-STE, needing less labeled triplets and producing higher quality embeddings.Description
Supervisor
Gionis, AristidesThesis advisor
Ukkonen, AnttiKeywords
embeddings, representation learning, ordinal constraints, relative similarity comparisons, probabilistic triplets, certainty