微调多层自编码算法

From Ufldl

Jump to: navigation, search
Line 7: Line 7:
Wiki上传者:崔巍,email:watsoncui@gmail.com,新浪微博:@太二真人
Wiki上传者:崔巍,email:watsoncui@gmail.com,新浪微博:@太二真人
-
:【原文】:
 
=== Introduction ===
=== Introduction ===
 +
:【原文】:
Fine tuning is a strategy that is commonly found in deep learning. As such, it can also be used to greatly improve the performance of a stacked autoencoder. From a high level perspective, fine tuning treats all layers of a stacked autoencoder as a single model, so that in one iteration, we are improving upon all the weights in the stacked autoencoder.
Fine tuning is a strategy that is commonly found in deep learning. As such, it can also be used to greatly improve the performance of a stacked autoencoder. From a high level perspective, fine tuning treats all layers of a stacked autoencoder as a single model, so that in one iteration, we are improving upon all the weights in the stacked autoencoder.
-
:【初译】:
+
:【初译】:简介
-
=== 简介 ===
+
微调是深度学习中的常用策略,可以大幅提升多层自动编码机的性能表现。从更高视角来说,微调将多层自动编码机的所有层视为一个模型,这样在每次迭代中,它所有权重值都可以被充分利用。
微调是深度学习中的常用策略,可以大幅提升多层自动编码机的性能表现。从更高视角来说,微调将多层自动编码机的所有层视为一个模型,这样在每次迭代中,它所有权重值都可以被充分利用。
Line 18: Line 17:
 +
=== General Strategy ===
:【原文】:
:【原文】:
-
=== General Strategy ===
 
Fortunately, we already have all the tools necessary to implement fine tuning for stacked autoencoders! In order to compute the gradients for all the layers of the stacked autoencoder in each iteration, we use the [[Backpropagation Algorithm]], as discussed in the sparse autoencoder section. As the backpropagation algorithm can be extended to apply for an arbitrary number of layers, we can actually use this algorithm on a stacked autoencoder of arbitrary depth.
Fortunately, we already have all the tools necessary to implement fine tuning for stacked autoencoders! In order to compute the gradients for all the layers of the stacked autoencoder in each iteration, we use the [[Backpropagation Algorithm]], as discussed in the sparse autoencoder section. As the backpropagation algorithm can be extended to apply for an arbitrary number of layers, we can actually use this algorithm on a stacked autoencoder of arbitrary depth.
-
:【初译】:
+
:【初译】:一般策略
-
=== 一般策略 ===
+
幸运的是,实现微调多层自动编码机的所有工具齐备。为了在每次迭代中对所有的层计算梯度,需要使用稀疏自编码一节讨论的反向传播算法。因为反向传播算法可以延伸应用到任意多层,所以事实上它对任意多层的自动编码机都适用。
幸运的是,实现微调多层自动编码机的所有工具齐备。为了在每次迭代中对所有的层计算梯度,需要使用稀疏自编码一节讨论的反向传播算法。因为反向传播算法可以延伸应用到任意多层,所以事实上它对任意多层的自动编码机都适用。
:【一审】:
:【一审】:
-
 
+
=== Finetuning with Backpropagation ===
:【原文】:
:【原文】:
-
=== Finetuning with Backpropagation ===
 
For your convenience, the summary of the backpropagation algorithm using element wise notation is below:
For your convenience, the summary of the backpropagation algorithm using element wise notation is below:
-
:【初译】:
+
:【初译】:使用反向传播微调
-
=== 使用反向传播微调 ===
+
为方便使用,以下简要描述了反向传播算法的使用:
为方便使用,以下简要描述了反向传播算法的使用:
:【一审】:
:【一审】:

Revision as of 18:27, 9 March 2013

Personal tools