Bonsai: A Framework for Convolutional Neural Network Acceleration Using Criterion-Based Pruning

TL;DR

This paper introduces Bonsai, a flexible framework for CNN pruning based on various criteria, enabling effective model compression and acceleration while maintaining or improving accuracy.

Contribution

It presents Combine, a novel criterion-based pruning framework with a standard comparison language and new criterion functions, improving CNN efficiency.

Findings

Pruned up to 79% of filters with maintained or improved accuracy.

Reduced network computations by up to 68%.

Demonstrated effectiveness on VGG-inspired models.

Abstract

As the need for more accurate and powerful Convolutional Neural Networks (CNNs) increases, so too does the size, execution time, memory footprint, and power consumption. To overcome this, solutions such as pruning have been proposed with their own metrics and methodologies, or criteria, for how weights should be removed. These solutions do not share a common implementation and are difficult to implement and compare. In this work, we introduce Combine, a criterion- based pruning solution and demonstrate that it is fast and effective framework for iterative pruning, demonstrate that criterion have differing effects on different models, create a standard language for comparing criterion functions, and propose a few novel criterion functions. We show the capacity of these criterion functions and the framework on VGG inspired models, pruning up to 79\% of filters while retaining or improving accuracy, and reducing the computations needed by the network by up to 68\%.