Learning Coarse-to-Fine Pruning of Graph Convolutional Networks for Skeleton-based Recognition

TL;DR

This paper introduces a novel coarse-to-fine pruning method for graph convolutional networks that combines the benefits of structured and unstructured pruning, leading to efficient and accurate skeleton-based recognition models.

Contribution

The proposed CTF pruning method models connection masks with a novel parametrization enabling flexible coarse-to-fine pruning, improving efficiency and accuracy in skeleton-based recognition.

Findings

Outperforms baseline pruning methods on SBU and FPHA datasets.

Achieves high efficiency without sacrificing accuracy.

Demonstrates effectiveness of coarse-to-fine pruning in GCNs.

Abstract

Magnitude Pruning is a staple lightweight network design method which seeks to remove connections with the smallest magnitude. This process is either achieved in a structured or unstructured manner. While structured pruning allows reaching high efficiency, unstructured one is more flexible and leads to better accuracy, but this is achieved at the expense of low computational performance. In this paper, we devise a novel coarse-to-fine (CTF) method that gathers the advantages of structured and unstructured pruning while discarding their inconveniences to some extent. Our method relies on a novel CTF parametrization that models the mask of each connection as the Hadamard product involving four parametrizations which capture channel-wise, column-wise, row-wise and entry-wise pruning respectively. Hence, fine-grained pruning is enabled only when the coarse-grained one is disabled, and this leads to highly efficient networks while being effective. Extensive experiments conducted on the challenging task of skeleton-based recognition, using the standard SBU and FPHA datasets, show the clear advantage of our CTF approach against different baselines as well as the related work.