Batch_Normalization: Learning rate 更大也没关系
发布人
打开封面 下载高清视频 观看高清视频 视频下载器
...
Batch_Normalization: Abstract
...
Batch_Normalization: Introduction
...
Batch_Normalization: Convolutional Network
...
Batch_Normalization: Toward Reducing Internal Covariance Shift
...
Batch_Normalization: Normalization vs mini-batch statistics
...
Dropout: 如何理解weights * (1- dropout_rate)
...
李宏毅深度学习batch normalization笔记要点问题
...
Batch_Normalization: Normalization vs mini-batch statistics full
...
(中文) 掰开揉碎：initialization, covariance shift, batch normalization
...
论文分解器：Generalization in deep learning (合集）
...
为什么当learning_rate 变大如>= 0.01 时，loss为NaN？
...
resnet 3.4 implementation
...
3.1 residual learning
...
Inception: architecture details
...
AlignedReID：如何理解mutual learning
...
ResNet related works
...
AlignedReID：mutual learning 效果的实验对比
...
Batch Normalization at test time
...
Dropout: motivation 2
...
Dropout: abstract
...
Inception: related works
...
ResNet: abstract and introduction
...
EN 掰开揉碎：initialization, batch normalization, covariance shift
...
Dropout: 4 model description
...
Dropout: 5.1 backpropagation
...
Dropout: Introduction
...
Dropout: Motivation 01
...
Dropout: related works
...
Inception: Abstract and Introduction
...
AlignedReID：triHard, triplet loss, 为什么在预测时只用global features
...
YOLO：abstract
...
论文分解器：medium's capsule 2.1
...
YOLO：Unified Detection part2
...
Stories of infection
...
(中文) 论文分解器：Hinton's capsule section7-8
...
论文分解器：medium's capsule 2.4
...
论文分解器：hinton's capsule 01 中文
...
(中文) 论文分解器：Hinton's capsule 2 paragraph1-4
...
论文分解器：medium's capsule 1.1
...
哈佛通识概率普及课（特别推荐）Fat Chance: Probability from the Ground Up by Harvard and Edx