Learning Centre

Learning Embeddings from Probabilistic Triplet Comparisons

 |  Login

Show simple item record

dc.contributor Aalto-yliopisto fi
dc.contributor Aalto University en
dc.contributor.advisor Ukkonen, Antti
dc.contributor.author Mojsilovic, Stefan
dc.date.accessioned 2018-10-17T08:06:24Z
dc.date.available 2018-10-17T08:06:24Z
dc.date.issued 2018-10-08
dc.identifier.uri https://aaltodoc.aalto.fi/handle/123456789/34375
dc.description.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. en
dc.format.extent 79+2
dc.language.iso en en
dc.title Learning Embeddings from Probabilistic Triplet Comparisons en
dc.type G2 Pro gradu, diplomityö fi
dc.contributor.school Perustieteiden korkeakoulu fi
dc.subject.keyword embeddings en
dc.subject.keyword representation learning en
dc.subject.keyword ordinal constraints en
dc.subject.keyword relative similarity comparisons en
dc.subject.keyword probabilistic triplets en
dc.subject.keyword certainty en
dc.identifier.urn URN:NBN:fi:aalto-201810175450
dc.programme.major Networking Technology en
dc.programme.mcode ELEC3029 fi
dc.type.ontasot Master's thesis en
dc.type.ontasot Diplomityö fi
dc.contributor.supervisor Gionis, Aristides
dc.programme Master’s Programme in Computer, Communication and Information Sciences fi
local.aalto.electroniconly yes
local.aalto.openaccess no

Files in this item

Files Size Format View

There are no open access files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search archive

Advanced Search

article-iconSubmit a publication