鹿鹿的博客

Reference https://zhuanlan.zhihu.com/p/32626442 https://zhuanlan.zhihu.com/p/21360434?refer=intelligentunit 深度学习优化算法经历了 SGD -> SGDM -> NAG ->AdaGrad -> AdaDelta -> Adam -> Nadam 这样的发...

Reference https://yogayu.github.io/DeepLearningCourse/03/ActivateFunction.html Deep Learning v1 笔记 https://mp.weixin.qq.com/s/7DgiXCNBS5vb07WIKTFYRQ 可视化神经网络中的激活函数——26 个激活函数的图示及其一阶导数 Recall that we use...

GoogLeNet（Inception-v1） 相比AlexNet和VGG，出现了多支路，引入了1×1卷积帮助减少网络计算量 Inception-v2 引入Batch Normalization(BN)；5×5卷积使用两个3×3卷积代替 Inception-v3 非对称卷积（n×n卷积分割为1×n和n×1两个卷积）; new pooling （为了防止信息丢失且不增加计算量，把串联的p...

VGG 2014 Very Deep Convolutional Networks for Large-Scale Image Recognition # ResNet 2015 [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) Residual Representations /...