Investigating the Not-So-Obvious Effects of Structured Pruning

Hugo Tessier (1) , Vincent Gripon (2, 3) , Mathieu Léonardon (2, 3) , Matthieu Arzel (2, 3) , David Bertrand (1) , Thomas Hannagan (4)

1 Stellantis - PSA Centre Technique de Vélizy
2 Lab-STICC_2AI - Equipe Algorithm Architecture Interactions
3 IMT Atlantique - MEE - Département Mathematical and Electrical Engineering
4 Stellantis France

Structured pruning is a popular method to reduce the cost of convolutional neural networks. However, depending on the architecture, pruning introduces dimensional discrepancies which prevent the actual reduction of pruned networks and mask their true complexity. Most papers in the literature overlook these issues. We propose a method that systematically solves them and generate an operational network. We show through experiments the gap between the theoretical pruning ratio and the actual complexity revealed by our method.

Domaines: Intelligence artificielle [cs.AI]

Hugo Tessier, Vincent Gripon, Mathieu Léonardon, Matthieu Arzel, David Bertrand, et al.. Investigating the Not-So-Obvious Effects of Structured Pruning. ICML 2022 - Hardware-aware efficient training (HAET), Jul 2022, Baltimore, United States. ⟨hal-03706472⟩