# Are Bias Mitigation Techniques for Deep Learning Effective?

arXiv: https://arxiv.org/abs/2104.00170 · HTML: https://mlsanity.com/paper/2104.00170/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 49 of its 75 references that are in the map.
- First posted on arXiv 1 Apr 2021.

## Builds on (most cited first, 30 of 49)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering](https://mlsanity.com/paper/1707.07998/) (2017)
- [Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization](https://mlsanity.com/paper/1911.08731/) (2019)
- [Class-Balanced Loss Based on Effective Number of Samples](https://mlsanity.com/paper/1901.05555/) (2019)
- [Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels](https://mlsanity.com/paper/1805.07836/) (2018)
- [Out-of-Distribution Generalization via Risk Extrapolation (REx)](https://mlsanity.com/paper/2003.00688/) (2020)
- [Fair prediction with disparate impact: A study of bias in recidivism prediction instruments](https://mlsanity.com/paper/1610.07524/) (2016)
- [Mitigating Unwanted Biases with Adversarial Learning](https://mlsanity.com/paper/1801.07593/) (2018)
- [Causal inference using invariant prediction: identification and confidence intervals](https://mlsanity.com/paper/1501.01332/) (2015)
- [Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints](https://mlsanity.com/paper/1707.09457/) (2017)
- [The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision](https://mlsanity.com/paper/1904.12584/) (2019)
- [The Pitfalls of Simplicity Bias in Neural Networks](https://mlsanity.com/paper/2006.07710/) (2020)
- [An Investigation of Why Overparameterization Exacerbates Spurious Correlations](https://mlsanity.com/paper/2005.04345/) (2020)
- [An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution](https://mlsanity.com/paper/1807.03247/) (2018)
- [Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding](https://mlsanity.com/paper/1810.02338/) (2018)
- [Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases](https://mlsanity.com/paper/1909.03683/) (2019)
- [Gradient Starvation: A Learning Proclivity in Neural Networks](https://mlsanity.com/paper/2011.09468/) (2020)
- [The Risks of Invariant Risk Minimization](https://mlsanity.com/paper/2010.05761/) (2020)
- [Learning Not to Learn: Training Deep Neural Networks with Biased Data](https://mlsanity.com/paper/1812.10352/) (2018)
- [Learning De-biased Representations with Biased Representations](https://mlsanity.com/paper/1910.02806/) (2019)
- [RUBi: Reducing Unimodal Biases in Visual Question Answering](https://mlsanity.com/paper/1906.10169/) (2019)
- [Statistics of Robust Optimization: A Generalized Empirical Likelihood Approach](https://mlsanity.com/paper/1610.03425/) (2016)
- [REPAIR: Removing Representation Bias by Dataset Resampling](https://mlsanity.com/paper/1904.07911/) (2019)
- [Learning from Failure: Training Debiased Classifier from Biased Classifier](https://mlsanity.com/paper/2007.02561/) (2020)
- [Taking a HINT: Leveraging Explanations to Make Vision and Language Models More Grounded](https://mlsanity.com/paper/1902.03751/) (2019)
- [What shapes feature representations? Exploring datasets, architectures, and training](https://mlsanity.com/paper/2006.12433/) (2020)
- [Unlearn Dataset Bias in Natural Language Inference by Fitting the Residual](https://mlsanity.com/paper/1908.10763/) (2019)
- [Towards Debiasing NLU Models from Unknown Biases](https://mlsanity.com/paper/2009.12303/) (2020)

## Built on by (most cited first, 14 of 14)

- [A Whac-A-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Others](https://mlsanity.com/paper/2212.04825/) (2022)
- [Discover and Mitigate Unknown Biases with Debiasing Alternate Networks](https://mlsanity.com/paper/2207.10077/) (2022)
- [Mitigating Dataset Bias by Using Per-sample Gradient](https://mlsanity.com/paper/2205.15704/) (2022)
- [Men Also Do Laundry: Multi-Attribute Bias Amplification](https://mlsanity.com/paper/2210.11924/) (2022)
- [OccamNets: Mitigating Dataset Bias by Favoring Simpler Hypotheses](https://mlsanity.com/paper/2204.02426/) (2022)
- [Supervising the Transfer of Reasoning Patterns in VQA](https://mlsanity.com/paper/2106.05597/) (2021)
- [General Greedy De-bias Learning](https://mlsanity.com/paper/2112.10572/) (2021)
- [Learning Generalizable Models via Disentangling Spurious and Enhancing Potential Correlations](https://mlsanity.com/paper/2401.05752/) (2024)
- [ChiQA: A Large Scale Image-based Real-World Question Answering Dataset for Multi-Modal Understanding](https://mlsanity.com/paper/2208.03030/) (2022)
- [Human in the Latent Loop (HILL): Interactively Guiding Model Training Through Human Intuition](https://mlsanity.com/paper/2505.06325/) (2025)
- [Dataverse: Open-Source ETL (Extract, Transform, Load) Pipeline for Large Language Models](https://mlsanity.com/paper/2403.19340/) (2024)
- [Implicit Visual Bias Mitigation by Posterior Estimate Sharpening of a Bayesian Neural Network](https://mlsanity.com/paper/2303.16564/) (2023)
- [Revisiting the Dataset Bias Problem from a Statistical Perspective](https://mlsanity.com/paper/2402.03577/) (2024)
- [Beyond Performance Disparities: A Three-Level Audit of Representational Harm in CelebA](https://mlsanity.com/paper/2605.15312/) (2026)

## Abstract (arXiv)

A critical problem in deep learning is that systems learn inappropriate biases, resulting in their inability to perform well on minority groups. This has led to the creation of multiple algorithms that endeavor to mitigate bias. However, it is not clear how effective these methods are. This is because study protocols differ among papers, systems are tested on datasets that fail to test many forms of bias, and systems have access to hidden knowledge or are tuned specifically to the test set. To address this, we introduce an improved evaluation protocol, sensible metrics, and a new dataset, which enables us to ask and answer critical questions about bias mitigation algorithms. We evaluate seven state-of-the-art algorithms using the same network architecture and hyperparameter selection policy across three benchmark datasets. We introduce a new dataset called Biased MNIST that enables assessment of robustness to multiple bias sources. We use Biased MNIST and a visual question answering (VQA) benchmark to assess robustness to hidden biases. Rather than only tuning to the test set distribution, we study robustness across different tuning distributions, which is critical because for many applications the test distribution may not be known during development. We find that algorithms exploit hidden biases, are unable to scale to multiple forms of bias, and are highly sensitive to the choice of tuning set. Based on our findings, we implore the community to adopt more rigorous assessment of future bias mitigation methods. All data, code, and results are publicly available at: https://github.com/erobic/bias-mitigators.