2026-09-19T19:50:43Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1390792022-01-15T03:34:13Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Agrawal, Pulkit
Simonovikj, Sanja
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2022-01-14T14:48:42Z
2022-01-14T14:48:42Z
2021-06
2021-06-17T20:14:23.018Z
https://hdl.handle.net/1721.1/139079
Deep Neural Networks (DNNs) find one out of many possible solutions to a given task such as classification. This solution is more likely to pick up on spurious features and low-level statistical patterns in the train data rather than semantic features and highlevel abstractions, resulting in poor Out-of-Distribution (OOD) performance. In this project we aim to broaden the current knowledge surrounding spurious correlations as they relate to DNNs. We do this by measuring their effect on generalization under various settings, determining the existence of subnetworks in a DNN that capture the core features and examine potential mitigation strategies. Finally, we discuss alternative approaches that are reserved for future work.
M.Eng.
Massachusetts Institute of Technology
In Copyright - Educational Use Permitted
Copyright MIT
http://rightsstatements.org/page/InC-EDU/1.0/
Towards Understanding Human-aligned Neural Representation in the Presence of Confounding Variables
Thesis
application/pdf
Master
Master of Engineering in Electrical Engineering and Computer Science
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="7a9c4c4a-28d1-4946-9aa7-988129771249">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Towards Understanding Human-aligned Neural Representation in the Presence of Confounding Variables</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2021-06</PublicationDate>
<Authors>
<Author>
<DisplayName>Simonovikj, Sanja</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://rightsstatements.org/page/InC-EDU/1.0/</License>
<Abstract>Deep Neural Networks (DNNs) find one out of many possible solutions to a given task such as classification. This solution is more likely to pick up on spurious features and low-level statistical patterns in the train data rather than semantic features and highlevel abstractions, resulting in poor Out-of-Distribution (OOD) performance. In this project we aim to broaden the current knowledge surrounding spurious correlations as they relate to DNNs. We do this by measuring their effect on generalization under various settings, determining the existence of subnetworks in a DNN that capture the core features and examine potential mitigation strategies. Finally, we discuss alternative approaches that are reserved for future work.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>