2026-09-19T07:28:50Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1307172026-06-06T00:48:58Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Antonio Torralba and David Bau.
You, Yejin.
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2021-05-24T19:52:49Z
2021-05-24T19:52:49Z
2021
2021
https://hdl.handle.net/1721.1/130717
1251801865
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, February, 2021
Cataloged from the official PDF of thesis.
Includes bibliographical references (pages 61-62).
In this thesis, we compare the representations of an unsupervised contrastive model to those of an equivalent supervised model using several deep neural network interpretability methods: network dissection, sparsity experiments, and saliency maps. Network dissections of self-supervised contrastive and supervised models show that the neurons of the contrastive model tend to learn about different parts of an object (ie. top-half of a dog or left-half of a person) while the neurons of the supervised model tend to learn about the entire object (ie. a dog or a person). Sparsity experiments show that the representations learned by the contrastive model are less sparse than the representations learned by the supervised counterpart model. Saliency maps show that the contrastive model focuses more on specific parts of the input image. Finally, we find that the contrastive model representations transfer better to finegrained classification tasks than the supervised model representations.
by Yejin You.
M.Eng.
M.Eng. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science
62 pages
eng
Massachusetts Institute of Technology
MIT theses may be protected by copyright. Please reuse MIT thesis content according to the MIT Libraries Permissions Policy, which is available through the URL provided.
http://dspace.mit.edu/handle/1721.1/7582
Electrical Engineering and Computer Science.
Contrasting contrastive and supervised models interpretability
Thesis
application/pdf
2021-05-24T19:52:49Z
Publication
Master
EECS
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="2047ac35-95ac-4725-81a5-84de1965372b">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>eng</Language>
<Title>Contrasting contrastive and supervised models interpretability</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2021</PublicationDate>
<Authors>
<Author>
<DisplayName>You, Yejin.</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://dspace.mit.edu/handle/1721.1/7582</License>
<Keyword>Electrical Engineering and Computer Science.</Keyword>
<Abstract>In this thesis, we compare the representations of an unsupervised contrastive model to those of an equivalent supervised model using several deep neural network interpretability methods: network dissection, sparsity experiments, and saliency maps. Network dissections of self-supervised contrastive and supervised models show that the neurons of the contrastive model tend to learn about different parts of an object (ie. top-half of a dog or left-half of a person) while the neurons of the supervised model tend to learn about the entire object (ie. a dog or a person). Sparsity experiments show that the representations learned by the contrastive model are less sparse than the representations learned by the supervised counterpart model. Saliency maps show that the contrastive model focuses more on specific parts of the input image. Finally, we find that the contrastive model representations transfer better to finegrained classification tasks than the supervised model representations.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>