<?xml version="1.0"?>
<?xml-stylesheet type="text/css" href="http://index.cslt.org/mediawiki/skins/common/feed.css?303"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="zh-cn">
		<id>http://index.cslt.org/mediawiki/index.php?action=history&amp;feed=atom&amp;title=Chao_Xing_2015-09-28</id>
		<title>Chao Xing 2015-09-28 - 版本历史</title>
		<link rel="self" type="application/atom+xml" href="http://index.cslt.org/mediawiki/index.php?action=history&amp;feed=atom&amp;title=Chao_Xing_2015-09-28"/>
		<link rel="alternate" type="text/html" href="http://index.cslt.org/mediawiki/index.php?title=Chao_Xing_2015-09-28&amp;action=history"/>
		<updated>2026-04-09T04:51:36Z</updated>
		<subtitle>本wiki的该页面的版本历史</subtitle>
		<generator>MediaWiki 1.23.3</generator>

	<entry>
		<id>http://index.cslt.org/mediawiki/index.php?title=Chao_Xing_2015-09-28&amp;diff=16925&amp;oldid=prev</id>
		<title>Xingchao：以“Last Week:  Last Week Solution:    1. DNN program has some problems, fix these &amp; DNN contain same performance to Linear.       Problem is training process, training...”为内容创建页面</title>
		<link rel="alternate" type="text/html" href="http://index.cslt.org/mediawiki/index.php?title=Chao_Xing_2015-09-28&amp;diff=16925&amp;oldid=prev"/>
				<updated>2015-09-28T02:30:58Z</updated>
		
		<summary type="html">&lt;p&gt;以“Last Week:  Last Week Solution:    1. DNN program has some problems, fix these &amp;amp; DNN contain same performance to Linear.       Problem is training process, training...”为内容创建页面&lt;/p&gt;
&lt;p&gt;&lt;b&gt;新页面&lt;/b&gt;&lt;/p&gt;&lt;div&gt;Last Week:&lt;br /&gt;
 Last Week Solution:&lt;br /&gt;
   1. DNN program has some problems, fix these &amp;amp; DNN contain same performance to Linear.&lt;br /&gt;
      Problem is training process, training samples are equal to speakers' so this will lead model less compute 50% training samples.&lt;br /&gt;
 Plan to do:&lt;br /&gt;
   1. Help Lantian finish his papers.&lt;br /&gt;
   2. Run for a small vocabulary size, such as 5000.&lt;br /&gt;
 Problem:&lt;br /&gt;
   1. DNN still not better than linear model.&lt;br /&gt;
   &lt;br /&gt;
This Week:&lt;br /&gt;
  Solution:&lt;br /&gt;
   1. DNN with dropout method, and add SAA to reduce randomly properties.&lt;br /&gt;
  Plan to do:&lt;br /&gt;
   1. Run for a small vocabulary size, such as 5000.&lt;br /&gt;
   2. Run and test some ideas.&lt;/div&gt;</summary>
		<author><name>Xingchao</name></author>	</entry>

	</feed>