33
/fr/
AIzaSyAYiBZKx7MnpbEhh9jyipgxe19OcubqV5w
August 1, 2025
3251714
288434
1
Public Timelines
Public TimelinesFAQ
Public TimelinesFAQ
For educational institutions For teachers For students Cabinet
For educational institutions For teachers For students Open cabinet
Créer
Create a timeline
Public timelines
Library
FAQ

5 janv. 2017 - Deep Learning with Low Precision by Half-wave Gaussian Quantization (HWGQ)

Description:

本文主要讨论了在低位宽量化的近似梯度训练中，梯度消失的问题，通过设计新的量化函数(sign/Q)和激活函数来保证梯度不消失，以实现量化模型的收敛。有趣的是，仅通过对relu函数的剪切或者log-tail变换，模型的性能得到了显著提升。这或许就是特征的影响

Ajouté au bande de temps:

Quantization timeline

By龚成
29 mars 2020
1
0
745

Date:

5 janv. 2017
Maintenaint
~ Il y a 8 ans et 5 mois
About & FeedbackUn accordConfidentialitéBibliothèqueFAQ
Support 24/7  
CabinetGet premium
Donate
The service accepts bank transfer (ACH, Wire) or cards (Visa, MasterCard, etc). Processed by Stripe.
Secured with SSL