2026-09-19T16:56:11Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1433242022-06-16T03:37:53Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Rus, Daniela L.
Phillips, Jacob
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2022-06-15T13:12:39Z
2022-06-15T13:12:39Z
2022-02
2022-02-22T18:32:27.546Z
https://hdl.handle.net/1721.1/143324
Traditional training regimens for time-series models have been shown to encode the biases from their training corpora into the models themselves. We aim to train unbiased time-series models using existing biased datasets. However, most debiasing techniques rely on explicit labels that encapsulate the bias, such as pairs of words along some worrying axis of bias such as race or gender for language models. We propose an unsupervised latent debiasing training regimen based on [2] that simultaneously learns the latent distribution of the dataset and a separate language task; datapoints are selected for training batches by sampling weights inverse to their commonality as determined by their placement in the latent space. We adapt [2] to time-series datasets and show algorithmic improvements to bias identification and bias reduction for models trained on toy and real datasets.
M.Eng.
Massachusetts Institute of Technology
In Copyright - Educational Use Permitted
Copyright MIT
http://rightsstatements.org/page/InC-EDU/1.0/
Unsupervised Latent Debiasing of Time-Series Models
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="5688700a-7b45-451b-b3c7-7e0b582db41c">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Unsupervised Latent Debiasing of Time-Series Models</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2022-02</PublicationDate>
<Authors>
<Author>
<DisplayName>Phillips, Jacob</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>Traditional training regimens for time-series models have been shown to encode the biases from their training corpora into the models themselves. We aim to train unbiased time-series models using existing biased datasets. However, most debiasing techniques rely on explicit labels that encapsulate the bias, such as pairs of words along some worrying axis of bias such as race or gender for language models. We propose an unsupervised latent debiasing training regimen based on [2] that simultaneously learns the latent distribution of the dataset and a separate language task; datapoints are selected for training batches by sampling weights inverse to their commonality as determined by their placement in the latent space. We adapt [2] to time-series datasets and show algorithmic improvements to bias identification and bias reduction for models trained on toy and real datasets.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>