Translations:Convolutional Neural Networks/22/zh: Difference between revisions

Revision as of 21:58, 27 April 2026

Information about message (contribute)

This message has no documentation. If you know where or how this message is used, you can help other translators by adding documentation to this message.

Message definition (Convolutional Neural Networks)

{| class="wikitable"
|-
! Architecture !! Year !! Key contribution !! Depth
|-
| '''LeNet-5''' || 1998 || Pioneered CNNs for handwritten digit recognition (MNIST) || 5 layers
|-
| '''AlexNet''' || 2012 || Won ImageNet; popularised ReLU, {{Term|dropout}}, GPU training || 8 layers
|-
| '''VGGNet''' || 2014 || Showed depth matters; used only <math>3 \times 3</math> filters throughout || 16–19 layers
|-
| '''GoogLeNet (Inception)''' || 2014 || Introduced inception modules with parallel filter sizes || 22 layers
|-
| '''ResNet''' || 2015 || Introduced residual connections enabling very deep networks || 50–152+ layers
|-
| '''DenseNet''' || 2017 || Connected each layer to every subsequent layer via dense blocks || 121–264 layers
|-
| '''EfficientNet''' || 2019 || Compound scaling of depth, width, and resolution || Variable
|}

架构	年份	关键贡献	深度
LeNet-5	1998	开创了用于手写数字识别（MNIST）的 CNN	5 层
AlexNet	2012	赢得 ImageNet；推广了 ReLU、dropout 和 GPU 训练	8 层
VGGNet	2014	证明深度很重要；全程仅使用 $3 \times 3$ 滤波器	16–19 层
GoogLeNet (Inception)	2014	引入了具有并行滤波器尺寸的 inception 模块	22 层
ResNet	2015	引入残差连接，使非常深的网络成为可能	50–152+ 层
DenseNet	2017	通过密集块将每一层与所有后续层相连接	121–264 层
EfficientNet	2019	对深度、宽度和分辨率进行复合缩放	可变

@@ Line 7: / Line 7: @@
 | '''AlexNet''' || 2012 || 赢得 ImageNet；推广了 ReLU、dropout 和 GPU 训练 || 8 层
 |-
-| '''VGGNet''' || 2014 || 表明深度很重要；全程仅使用 <math>3 \times 3</math> 滤波器 || 16–19 层
+| '''VGGNet''' || 2014 || 证明深度很重要；全程仅使用 <math>3 \times 3</math> 滤波器 || 16–19 层
 |-
 | '''GoogLeNet (Inception)''' || 2014 || 引入了具有并行滤波器尺寸的 inception 模块 || 22 层
 |-
-| '''ResNet''' || 2015 || 引入残差连接，使训练非常深的网络成为可能 || 50–152+ 层
+| '''ResNet''' || 2015 || 引入残差连接，使非常深的网络成为可能 || 50–152+ 层
 |-
-| '''DenseNet''' || 2017 || 通过密集块将每一层与之后的所有层相连 || 121–264 层
+| '''DenseNet''' || 2017 || 通过密集块将每一层与所有后续层相连接 || 121–264 层
 |-
 | '''EfficientNet''' || 2019 || 对深度、宽度和分辨率进行复合缩放 || 可变
 |}