Multi-Dimensional Pruning: Joint Channel, Layer and Block Pruning with Latency Constraint

As we push the boundaries of performance in various vision tasks, the models grow in size correspondingly. To keep up with this growth, we need very aggressive pruning techniques for efficient inference and deployment on edge devices. Existing pruning approaches are limited to channel pruning and st...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2024-06
Hauptverfasser:	Sun, Xinglong, Barath Lakshmanan, Shen, Maying, Shiyi Lan, Chen, Jingde, Alvarez, Jose
Format:	Artikel
Sprache:	eng
Schlagworte:	Accuracy Mixed integer Nonlinear programming Object recognition Pruning
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!