Pruning by Explaining: A Novel Criterion for Deep Neural Network Pruning

TL;DR

This paper introduces a neural network pruning method based on explainability scores, effectively reducing model size while maintaining or improving accuracy, especially in low-data transfer learning scenarios.

Contribution

The paper presents a novel explainability-inspired criterion for CNN pruning that outperforms existing methods in resource-constrained transfer learning settings.

Findings

Outperforms state-of-the-art pruning criteria with retraining.

Effectively prunes models in low-data transfer scenarios.

Maintains or improves accuracy after pruning.

Abstract

The success of convolutional neural networks (CNNs) in various applications is accompanied by a significant increase in computation and parameter storage costs. Recent efforts to reduce these overheads involve pruning and compressing the weights of various layers while at the same time aiming to not sacrifice performance. In this paper, we propose a novel criterion for CNN pruning inspired by neural network interpretability: The most relevant units, i.e. weights or filters, are automatically found using their relevance scores obtained from concepts of explainable AI (XAI). By exploring this idea, we connect the lines of interpretability and model compression research. We show that our proposed method can efficiently prune CNN models in transfer-learning setups in which networks pre-trained on large corpora are adapted to specialized tasks. The method is evaluated on a broad range of computer vision datasets. Notably, our novel criterion is not only competitive or better compared to state-of-the-art pruning criteria when successive retraining is performed, but clearly outperforms these previous criteria in the resource-constrained application scenario in which the data of the task to be transferred to is very scarce and one chooses to refrain from fine-tuning. Our method is able to compress the model iteratively while maintaining or even improving accuracy. At the same time, it has a computational cost in the order of gradient computation and is comparatively simple to apply without the need for tuning hyperparameters for pruning.