# TARA: Training and Representation Alteration for AI Fairness and Domain Generalization

arXiv: https://arxiv.org/abs/2012.06387 · HTML: https://mlsanity.com/paper/2012.06387/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Dec 2020.

## Built on by (most cited first, 10 of 10)

- [FairREAD: Re-fusing Demographic Attributes after Disentanglement for Fair Medical Image Classification](https://mlsanity.com/paper/2412.16373/) (2024)
- [EdgeMixup: Improving Fairness for Skin Disease Classification and Segmentation](https://mlsanity.com/paper/2202.13883/) (2022)
- [Renyi Fair Information Bottleneck for Image Classification](https://mlsanity.com/paper/2203.04950/) (2022)
- [Detecting and Mitigating Algorithmic Bias in Binary Classification using Causal Modeling](https://mlsanity.com/paper/2310.12421/) (2023)
- [Adaptation and Generalization for Unknown Sensitive Factors of Variations](https://mlsanity.com/paper/2107.13625/) (2021)
- [Classification Utility, Fairness, and Compactness via Tunable Information Bottleneck and R'enyi Measures](https://mlsanity.com/paper/2206.10043/) (2022)
- [Evaluating Trade-offs in Computer Vision Between Attribute Privacy, Fairness and Utility](https://mlsanity.com/paper/2302.07917/) (2023)
- [Robustness and Adaptation to Hidden Factors of Variation](https://mlsanity.com/paper/2203.01864/) (2022)
- [Survey of Trustworthy AI: A Meta Decision of AI](https://mlsanity.com/paper/2306.00380/) (2023)
- [Mitigating Algorithmic Bias in Multiclass CNN Classifications Using Causal Modeling](https://mlsanity.com/paper/2501.07885/) (2025)

## Abstract (arXiv)

We propose a novel method for enforcing AI fairness with respect to protected or sensitive factors. This method uses a dual strategy performing training and representation alteration (TARA) for the mitigation of prominent causes of AI bias by including: a) the use of representation learning alteration via adversarial independence to suppress the bias-inducing dependence of the data representation from protected factors; and b) training set alteration via intelligent augmentation to address bias-causing data imbalance, by using generative models that allow the fine control of sensitive factors related to underrepresented populations via domain adaptation and latent space manipulation. When testing our methods on image analytics, experiments demonstrate that TARA significantly or fully debiases baseline models while outperforming competing debiasing methods that have the same amount of information, e.g., with (% overall accuracy, % accuracy gap) = (78.8, 0.5) vs. the baseline method's score of (71.8, 10.5) for EyePACS, and (73.7, 11.8) vs. (69.1, 21.7) for CelebA. Furthermore, recognizing certain limitations in current metrics used for assessing debiasing performance, we propose novel conjunctive debiasing metrics. Our experiments also demonstrate the ability of these novel metrics in assessing the Pareto efficiency of the proposed methods.