Sep 13, 2023 · We develop a pairwise mixup scheme to augment training data and encourage fair and accurate decision boundaries for all subgroups.
In this work, we propose data augmentation via pairwise mixup across subgroups to improve group fairness. Many real-world ap- plications of machine learning ...
In particular, data augmentation plays a critical role in generating novel use cases for model evaluation, as in the case of perturbation analyses [85,89,107], ...
To improve the generalizability of fair classifiers, we propose fair mixup, a new data augmentation strategy for imposing the fairness constraint.
People also ask
What is mixup data augmentation?
How does data augmentation improve accuracy?
How does data augmentation reduce overfitting?
Does data augmentation improve generalization?
Mixup is a data augmentation method to create new training data by linearly interpolating between pairs of data samples and their labels. Clustering · Data ...
Jan 24, 2024 · GBMix, a group-balanced Mixup strategy to train fair classifiers. It groups the dataset based on their attributes and balances the Mixup ratio between the ...
Missing: Subgroup | Show results with:Subgroup
Oct 2, 2024 · We found that using linear mixup alone, a data augmentation technique, for bias mitigation, can still retain biases present in dataset labels.
Abstract. Data augmentation with Mixup has been proven an effec- tive method to regularize the current deep neural networks. Mixup generates virtual samples ...
Missing: Subgroup Fairness.
In this work, we propose data augmentation via pairwise mixup across subgroups to improve group fairness. Many real-world applications of machine learning ...
Jan 12, 2021 · To improve the generalizability of fair classifiers, we propose fair mixup, a new data augmentation strategy for imposing the fairness ...
Missing: Subgroup | Show results with:Subgroup