What is Left After Distillation? How Knowledge Transfer Impacts Fairness and Bias

TL;DR

This paper investigates how knowledge distillation affects class bias and fairness in deep neural networks, revealing that temperature tuning can improve fairness and individual fairness, with implications for sensitive applications.

Contribution

It demonstrates that distillation impacts class bias and fairness, and shows that adjusting temperature can enhance fairness metrics, sometimes surpassing the teacher model.

Findings

41% of classes are significantly affected by distillation.

Higher temperature improves fairness metrics like DPD and EOD.

Distilled models can outperform teachers in fairness at high temperatures.

Abstract

Knowledge Distillation is a commonly used Deep Neural Network (DNN) compression method, which often maintains overall generalization performance. However, we show that even for balanced image classification datasets, such as CIFAR-100, Tiny ImageNet and ImageNet, as many as 41% of the classes are statistically significantly affected by distillation when comparing class-wise accuracy (i.e. class bias) between a teacher/distilled student or distilled student/non-distilled student model. Changes in class bias are not necessarily an undesirable outcome when considered outside of the context of a model's usage. Using two common fairness metrics, Demographic Parity Difference (DPD) and Equalized Odds Difference (EOD) on models trained with the CelebA, Trifeature, and HateXplain datasets, our results suggest that increasing the distillation temperature improves the distilled student model's fairness, and the distilled student fairness can even surpass the fairness of the teacher model at high temperatures. Additionally, we examine individual fairness, ensuring similar instances receive similar predictions. Our results confirm that higher temperatures also improve the distilled student model's individual fairness. This study highlights the uneven effects of distillation on certain classes and its potentially significant role in fairness, emphasizing that caution is warranted when using distilled models for sensitive application domains.