Balanced Datasets Are Not Enough: Estimating and Mitigating Gender Bias in Deep Image Representations

  title={Balanced Datasets Are Not Enough: Estimating and Mitigating Gender Bias in Deep Image Representations},
  author={Tianlu Wang and Jieyu Zhao and Mark Yatskar and Kai-Wei Chang and Vicente Ordonez},
  journal={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
  • Tianlu Wang, Jieyu Zhao, +2 authors Vicente Ordonez
  • Published 2019
  • Computer Science
  • 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
  • In this work, we present a framework to measure and mitigate intrinsic biases with respect to protected variables --such as gender-- in visual recognition tasks. [...] Key Method To mitigate this, we adopt an adversarial approach to remove unwanted features corresponding to protected variables from intermediate representations in a deep neural network -- and provide a detailed analysis of its effectiveness. Experiments on two datasets: the COCO dataset (objects), and the imSitu dataset (actions), show…Expand Abstract
    41 Citations
    Representation Learning with Statistical Independence to Mitigate Bias.
    • 5
    • PDF
    Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation
    • 7
    • PDF
    Bias-Resilient Neural Network
    • 6
    Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
    • 48
    • PDF
    Debiasing Convolutional Neural Networks via Meta Orthogonalization
    Mitigating Gender Bias Amplification in Distribution by Posterior Regularization
    • 1
    • PDF


    Adversarial Removal of Demographic Attributes from Text Data
    • 86
    • PDF
    ConvNets and ImageNet Beyond Accuracy: Explanations, Bias Detection, Adversarial Examples and Model Criticism
    • 36
    • PDF
    Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints
    • 276
    • PDF
    Improving Smiling Detection with Race and Gender Diversity
    • 16
    • PDF
    Undoing the Damage of Dataset Bias
    • 303
    • PDF
    Mitigating Unwanted Biases with Adversarial Learning
    • 249
    • PDF
    Women also Snowboard: Overcoming Bias in Captioning Models
    • 119
    • PDF
    Towards Privacy-Preserving Visual Recognition via Adversarial Training: A Pilot Study
    • 51
    • PDF
    Controllable Invariance through Adversarial Feature Learning
    • 100
    • PDF