LEAN: graph-based pruning for convolutional neural networks by extracting longest chains

TL;DR

The paper introduces LEAN, a graph-based CNN pruning method that considers chains of operators, leading to significant filter reduction while maintaining accuracy across various image tasks.

Contribution

It proposes a novel graph-based pruning approach that extracts relevant operator chains, improving over filter-wise methods by accounting for interdependence.

Findings

Achieves 1.7-12x fewer filters with similar accuracy

Effective on multiple image-to-image tasks including medical imaging

Outperforms existing pruning approaches in filter reduction

Abstract

Neural network pruning techniques can substantially reduce the computational cost of applying convolutional neural networks (CNNs). Common pruning methods determine which convolutional filters to remove by ranking the filters individually, i.e., without taking into account their interdependence. In this paper, we advocate the viewpoint that pruning should consider the interdependence between series of consecutive operators. We propose the LongEst-chAiN (LEAN) method that prunes CNNs by using graph-based algorithms to select relevant chains of convolutions. A CNN is interpreted as a graph, with the operator norm of each operator as distance metric for the edges. LEAN pruning iteratively extracts the highest value path from the graph to keep. In our experiments, we test LEAN pruning on several image-to-image tasks, including the well-known CamVid dataset, and a real-world X-ray CT dataset. Results indicate that LEAN pruning can result in networks with similar accuracy, while using 1.7-12x fewer convolutional filters than existing approaches.