Bias in Pruned Vision Models: In-Depth Analysis and Countermeasures
This addresses bias issues in compressed models for computer vision applications, offering practical tools to predict and mitigate bias, though it is incremental as it builds on prior evidence.
The study investigated how pruning affects bias in vision models, finding that models can be pruned to less than 10% weights without losing accuracy or increasing bias, but higher sparsities lead to higher uncertainty and bias.
Pruning - that is, setting a significant subset of the parameters of a neural network to zero - is one of the most popular methods of model compression. Yet, several recent works have raised the issue that pruning may induce or exacerbate bias in the output of the compressed model. Despite existing evidence for this phenomenon, the relationship between neural network pruning and induced bias is not well-understood. In this work, we systematically investigate and characterize this phenomenon in Convolutional Neural Networks for computer vision. First, we show that it is in fact possible to obtain highly-sparse models, e.g. with less than 10% remaining weights, which do not decrease in accuracy nor substantially increase in bias when compared to dense models. At the same time, we also find that, at higher sparsities, pruned models exhibit higher uncertainty in their outputs, as well as increased correlations, which we directly link to increased bias. We propose easy-to-use criteria which, based only on the uncompressed model, establish whether bias will increase with pruning, and identify the samples most susceptible to biased predictions post-compression.