BriefGPT.xyz
Sep, 2016
通过子模性和超模性优化神经网络架构
Neural Network Architecture Optimization through Submodularity and Supermodularity
HTML
PDF
Junqi Jin, Ziang Yan, Kun Fu, Nan Jiang, Changshui Zhang
TL;DR
本文研究深度学习模型的架构对于模型测试精度和计算时间的影响，提出了一种基于子集选择问题的架构优化方法，通过实验验证了该方法的有效性并进行了架构演化的分析和设计建议。
Abstract
deep learning
models' architectures, including depth and width, are key factors influencing models' performance, such as test accuracy and
computation time
. This paper solves two problems: given
→