Representation Learning with Statistical Independence to Mitigate Bias
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Presence of bias (in datasets or tasks) is inarguably one of the most critical challenges in machine learning applications that has alluded to pivotal debates in recent years. Such challenges range from spurious associations between variables in medical studies to the bias of race in gender or face recognition systems. Controlling for all types of biases in the dataset curation stage is cumbersome and sometimes impossible. The alternative is to use the available data and build models incorporating fair representation learning. In this paper, we propose such a model based on adversarial training with two competing objectives to learn features that have (1) maximum discriminative power with respect to the task and (2) minimal statistical mean dependence with the protected (bias) variable(s). Our approach does so by incorporating a new adversarial loss function that encourages a vanished correlation between the bias and the learned features. We apply our method to synthetic data, medical images (containing task bias), and a dataset for gender classification (containing dataset bias). Our results show that the learned features by our method not only result in superior prediction performance but also are unbiased.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Baron, R. M., & Kenny, D. A. (1986) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.51.6.1173 · Cited by 91,532
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Self-citationDeep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
Simonyan, K., Vedaldi, A., & Zisserman, A. (2013) · arXiv · doi:10.48550/arxiv.1312.6034 · Cited by 5,370 · cited 1×
“For this purpose, we use a similar technique as in [45] to extract the pixels in the original image space highlighting the areas that are discriminative for the gender labels.”
MethodAdversarial Discriminative Domain Adaptation
Tzeng, E., Hoffman, J., Saenko, K., & Darrell, T. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.316 · Cited by 5,154 · cited 1×
“Several other works built on top of the same idea and explored different loss functions [9], domain discriminator settings [51, 17, 8], or cycle-consistency [25].”
BackgroundBarocas, S., & Selbst, A. D. (2014) · doi:10.2139/ssrn.2477899 · Cited by 3,048 · cited 1×
“It is often argued that human or society biases are replicated in the training datasets and hence can be seen in learned models [4].”
BackgroundMeasuring and testing dependence by correlation of distances
Székely, G. J., Rizzo, M. L., & Bakirov, N. K. (2007) · The Annals of Statistics · doi:10.1214/009053607000000505 · Cited by 2,684
CyCADA: Cycle-Consistent Adversarial Domain Adaptation
Hoffman, J., Tzeng, E., Park, T., Zhu, J.-Y., Isola, P., Saenko, K., Efros, A. A., & Darrell, T. (2017) · arXiv · doi:10.48550/arxiv.1711.03213 · Cited by 2,656 · cited 1×
“Several other works built on top of the same idea and explored different loss functions [9], domain discriminator settings [51, 17, 8], or cycle-consistency [25].”
BackgroundUnsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks
Bousmalis, K., Silberman, N., Dohan, D., Erhan, D., & Krishnan, D. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.18 · Cited by 1,603 · cited 1×
“Several other works built on top of the same idea and explored different loss functions [9], domain discriminator settings [51, 17, 8], or cycle-consistency [25].”
BackgroundMitigating Unwanted Biases with Adversarial Learning
Zhang, B. H., Lemoine, B., & Mitchell, M. (2018) · Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3278721.3278779 · Cited by 1,387
Cited by
Koçak, B., Ponsiglione, A., Stanzione, A., Bluethgen, C., Santinha, J., Ugga, L., Huisman, M., Klontzas, M. E., Cannella, R., & Cuocolo, R. (2024) · Diagnostic and Interventional Radiology · doi:10.4274/dir.2024.242854 · Cited by 168
Cary, M. P., Zink, A., Wei, S., Olson, A., Yan, M., Senior, R., Bessias, S., Gadhoumi, K., Jean-Pierre, G., Wang, D., et al. (2023) · Health Affairs · doi:10.1377/hlthaff.2023.00553 · Cited by 94
Confound Removal and Normalization in Practice: A Neuroimaging Based Sex Prediction Case Study
More, S., Eickhoff, S. B., Caspers, J., & Patil, K. R. (2021) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-67670-4_1 · Cited by 37
“Of note, confound removal can be seen as supplementary to debiasing and fair learning [18, 16, 2] but here we do not investigate this angle.”
Background
Metadata
- Authors
- Adeli, E., Zhao, Q., Pfefferbaum, A., Sullivan, E. V., Fei-Fei, L., Niebles, J. C., & Pohl, K. M.
- Venue
- 2021 IEEE Winter Conference on Applications of Computer Vision (WACV)
- Publisher
- IEEE
- Published
- 2021-01
- Volume
- 2021
- Pages
- 2512-2522
- ISSN
- 2472-6737 (ISSN-L)
- OpenAlex
- W3036785680
- Cited by
- 85
- References
- 15
- FWCI
- 16.23
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Representation Learning with Statistical Independence to Mitigate Bias. https://app.cochat.ai/papers/10.1109/wacv48630.2021.00256
- The paper
- Adeli, E., Zhao, Q., Pfefferbaum, A., Sullivan, E. V., Fei-Fei, L., Niebles, J. C., & Pohl, K. M. (2021). Representation Learning with Statistical Independence to Mitigate Bias. 2021 IEEE Winter Conference on Applications of Computer Vision (WACV). https://doi.org/10.1109/wacv48630.2021.00256
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.