Corpus ID: 232352329

Matched sample selection with GANs for mitigating attribute confounding

@article{Singh2021MatchedSS,
  title={Matched sample selection with GANs for mitigating attribute confounding},
  author={Chandan Singh and G. Balakrishnan and P. Perona},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.13455}
}
Measuring biases of vision systems with respect to protected attributes like gender and age is critical as these systems gain widespread use in society. However, significant correlations between attributes in benchmark datasets make it difficult to separate algorithmic bias from dataset bias. To mitigate such attribute confounding during bias analysis, we propose a matching [1] approach that selects a subset of images from the full dataset with balanced attribute distributions across protected… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 88 REFERENCES
Balanced Datasets Are Not Enough: Estimating and Mitigating Gender Bias in Deep Image Representations
  • 65
  • PDF
Towards causal benchmarking of bias in face analysis algorithms
  • 7
  • PDF
FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age
  • 45
  • PDF
REPAIR: Removing Representation Bias by Dataset Resampling
  • Y. Li, N. Vasconcelos
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2019
  • 51
  • PDF
Analyzing and Reducing the Damage of Dataset Bias to Face Recognition With Synthetic Data
  • 30
  • PDF
Empirically Analyzing the Effect of Dataset Biases on Deep Face Recognition Systems
  • 27
  • PDF
A Style-Based Generator Architecture for Generative Adversarial Networks
  • 1,903
  • PDF
The Counterfactual $\chi$-GAN
  • 1
  • PDF
Deep Learning Face Attributes in the Wild
  • 3,283
  • PDF
...
1
2
3
4
5
...