PAODING: A High-fidelity Data-free Pruning Toolkit for Debloating Pre-trained Neural Networks

TL;DR

PAODING is a data-free pruning toolkit that effectively reduces pretrained neural network sizes while maintaining high fidelity, generalizing across datasets and models, and preserving accuracy and robustness.

Contribution

It introduces a novel iterative, data-free neuron pruning method that preserves model performance and robustness, applicable to various models and datasets.

Findings

Significant model size reduction achieved

Maintains test accuracy and adversarial robustness

Effective across different datasets and models

Abstract

We present PAODING, a toolkit to debloat pretrained neural network models through the lens of data-free pruning. To preserve the model fidelity, PAODING adopts an iterative process, which dynamically measures the effect of deleting a neuron to identify candidates that have the least impact to the output layer. Our evaluation shows that PAODING can significantly reduce the model size, generalize on different datasets and models, and meanwhile preserve the model fidelity in terms of test accuracy and adversarial robustness. PAODING is publicly available on PyPI via https://pypi.org/project/paoding-dl.