BriefGPT.xyz
Mar, 2023
旋转不变量量化用于模型压缩
Rotation Invariant Quantization for Model Compression
HTML
PDF
Joseph Kampeas, Yury Nahshan, Hanoch Kremer, Gil Lederman, Shira Zaloshinski...
TL;DR
通过提出RIQ技术，我们证明了其在压缩神经网络模型方面的最优性，并在各种模型和任务上证明了其优异性能，包括在预训练的VGG dense和pruned模型上能导致高达x52.9的压缩比。
Abstract
Post-training
neural network
(NN)
model compression
is an attractive approach for deploying large, memory-consuming models on devices with limited memory resources. In this study, we investigate the
→