BAdd: Bias Mitigation through Bias Addition
Keywords: fairness, bias, spurious correlations
Abstract: Computer vision (CV) datasets often exhibit biases in the form of spurious correlations between certain attributes and target variables that are perpetuated by Deep Learning (DL) models. While recent efforts aim to mitigate such biases and foster bias-neutral representations, they fail in complex real-world scenarios. In particular, existing methods excel in controlled experiments on benchmarks with single-attribute injected biases, but struggle with complex multi-attribute biases that naturally occur in established CV datasets. Here, we introduce BAdd, a simple yet effective method that allows for learning bias-neutral representations invariant to bias-inducing attributes. It achieves this by injecting features encoding these attributes into the training process. BAdd is evaluated on seven benchmarks and exhibits competitive performance, surpassing state-of-the-art methods on both single- and multi-attribute bias settings. Notably, it achieves +27.5% and +5.5% absolute accuracy improvements on the challenging multi-attribute benchmarks, FB-Biased-MNIST and CelebA, respectively.
Supplementary Material: zip
Primary Area: alignment, fairness, safety, privacy, and societal considerations
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 6572
Loading