Skip to yearly menu bar Skip to main content


Recall Distortion in Neural Network Pruning and the Undecayed Pruning Algorithm

Aidan Good · Jiaqi Lin · Xin Yu · Hannah Sieg · Mikey Fergurson · Shandian Zhe · Jerzy Wieczorek · Thiago Serra

Hall J (level 1) #715


Pruning techniques have been successfully used in neural networks to trade accuracy for sparsity. However, the impact of network pruning is not uniform: prior work has shown that the recall for underrepresented classes in a dataset may be more negatively affected. In this work, we study such relative distortions in recall by hypothesizing an intensification effect that is inherent to the model. Namely, that pruning makes recall relatively worse for a class with recall below accuracy and, conversely, that it makes recall relatively better for a class with recall above accuracy. In addition, we propose a new pruning algorithm aimed at attenuating such effect. Through statistical analysis, we have observed that intensification is less severe with our algorithm but nevertheless more pronounced with relatively more difficult tasks, less complex models, and higher pruning ratios. More surprisingly, we conversely observe a de-intensification effect with lower pruning ratios.

Chat is not available.