MinMAE calibration method for convolutional neural network quantization

This article introduces MinMAE, a novel activation calibration method for Post-Training Quantization (PTQ) that significantly reduces accuracy loss in Convolutional Neural Networks (CNN). Motivated by the need for high-fidelity quantization without costly retraining, MinMAE directly minimizes the Me...

Descripció completa

Dades bibliogràfiques

Autors principals:	Artem Vasilev, Andrey I. Kapitanov, Roman Solovyev, Dmitry Telpukhov, Alexander Stempkovskiy, Alexander Kustov, Viktor Prutyanov, Aleksandr Romanov
Format:	Artigo
Idioma:	Inglês
Publicat:	PeerJ Inc. 2026-03-01
Col·lecció:	PeerJ Computer Science
Matèries:	Quantization Object detection Neural network
Accés en línia:	https://peerj.com/articles/cs-3724.pdf
Etiquetes:	Afegir etiqueta Sense etiquetes, Sigues el primer a etiquetar aquest registre!