Characterising Bias in Compressed Models

Pruning and quantization are widely applied techniques for compressing deep neural networks, often driven by the resource constraints of deploying models to mobile phones or embedded devices (Esteva et al., 2017; Lane & Warden, 2018). To-date, discussion around the relative merits of different compression methods has centered on the tradeoff between level of compression and top-line metrics such as top-1 and top-5 accuracy (Blalock et al., 2020). Along this dimension, compression techniques are remarkably successful. It is possible to prune the majority of weights (Gale et al., 2019; Evci et al., 2019) or heavily quantize the bit representation (Jacob et al., 2017) with negligible decreases to test-set accuracy. However, recent work by Hooker et al. (2019a) has found that the minimal changes to top-line metrics obscure critical differences in generalization between pruned and non-pruned networks. The authors establish that pruning disproportionately impacts predictive performance on a small subset of the dataset. We build upon this work and focus on the implications of these findings for a dataset with sensitive protected attributes such as gender and age. Our work addresses the question: Does compression amplify existing algorithmic bias?