重新思考卷积神经网络架构中各层特征数量
机器学习
2018-12-17 v1 计算机视觉与模式识别
机器学习
摘要
我们从每层相对特征数量的角度刻画卷积神经网络。以偏正态分布作为参数化框架，我们考察了架构设计中高层特征数单调增加的常见假设。在 MNIST、Fashion-MNIST 与 CIFAR-10 图像分类基准上基于 VGG 型层的模型评估提供了促使我们反思该常见假设的证据：偏好较大早期层的架构似乎能取得更好的准确率。
引用
@article{arxiv.1812.05836,
title = {Rethinking Layer-wise Feature Amounts in Convolutional Neural Network Architectures},
author = {Martin Mundt and Sagnik Majumder and Tobias Weis and Visvanathan Ramesh},
journal= {arXiv preprint arXiv:1812.05836},
year = {2018}
}
备注
Accepted at the Critiquing and Correcting Trends in Machine Learning (CRACT) Workshop at the 32nd Conference on Neural Information Processing Systems (NeurIPS 2018)