BriefGPT.xyz
Aug, 2023
准确神经网络剪枝需要重思稀疏优化
Accurate Neural Network Pruning Requires Rethinking Sparse Optimization
HTML
PDF
Denis Kuznedelev, Eldar Kurtic, Eugenia Iofinova, Elias Frantar, Alexandra Peste...
TL;DR
使用标准的计算机视觉和自然语言处理稀疏基准测试,探讨高稀疏性对模型训练的影响,提供了一种解决有关稀疏训练困难的新方法,并在高稀疏性环境下实现了在视觉模型和语言模型上最先进的结果。
Abstract
Obtaining versions of
deep neural networks
that are both highly-accurate and highly-sparse is one of the main challenges in the area of
model compression
, and several high-performance pruning techniques have been
→