Overwriting Pretrained Bias with Finetuning Data

Angelina Wang, Olga Russakovsky

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Transfer learning is beneficial by allowing the expressive features of models pretrained on large-scale datasets to be finetuned for the target task of smaller, more domain-specific datasets. However, there is a concern that these pretrained models may come with their own biases which would propagate into the finetuned model. In this work, we investigate bias when conceptualized as both spurious correlations between the target task and a sensitive attribute as well as underrepresentation of a particular group in the dataset. Under both notions of bias, we find that (1) models finetuned on top of pretrained models can indeed inherit their biases, but (2) this bias can be corrected for through relatively minor interventions to the finetuning dataset, and often with a negligible impact to performance. Our findings imply that careful curation of the finetuning dataset is important for reducing biases on a downstream task, and doing so can even compensate for bias in the pretrained model.

Original languageEnglish (US)
Title of host publicationProceedings - 2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages12
ISBN (Electronic)9798350307184
StatePublished - 2023
Event2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023 - Paris, France
Duration: Oct 2 2023Oct 6 2023

Publication series

NameProceedings of the IEEE International Conference on Computer Vision
ISSN (Print)1550-5499


Conference2023 IEEE/CVF International Conference on Computer Vision, ICCV 2023

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition


Dive into the research topics of 'Overwriting Pretrained Bias with Finetuning Data'. Together they form a unique fingerprint.

Cite this