原
Deep Learning 最优化方法之Adam
2017年05月21日 23:06:52
BVL10101111
阅读数:25548
<span class="tags-box artic-tag-box">
<span class="label">标签:</span>
<a data-track-click="{"mod":"popu_626","con":"深度学习"}" class="tag-link" href="http://so.csdn.net/so/search/s.do?q=深度学习&t=blog" target="_blank">深度学习 </a><a data-track-click="{"mod":"popu_626","con":"优化"}" class="tag-link" href="http://so.csdn.net/so/search/s.do?q=优化&t=blog" target="_blank">优化 </a>
<span class="article_info_click">收起</span></span>
<div class="tags-box space">
<span class="label">个人分类:</span>
<a class="tag-link" href="https://blog.csdn.net/BVL10101111/article/category/6546906" target="_blank">dl </a>
</div>
</div>
<div class="operating">
</div>
</div>
</div>
</div>
<article class="baidu_pl">
<div id="article_content" class="article_content clearfix csdn-tracking-statistics" data-pid="blog" data-mod="popu_307" data-dsm="post">
<div class="article-copyright">
版权声明:本文为博主原创文章,未经博主允许不得转载。 https://blog.csdn.net/BVL10101111/article/details/72616516 </div>
<div id="content_views" class="markdown_views prism-atom-one-dark">
<!-- flowchart 箭头图标 勿删 -->
<svg xmlns="http://www.w3.org/2000/svg" style="display: none;"><path stroke-linecap="round" d="M5,0 0,2.5 5,5z" id="raphael-marker-block" style="-webkit-tap-highlight-color: rgba(0, 0, 0, 0);"></path></svg>
<p>本文是<a href="http://blog.csdn.net/BVL10101111/article/details/72614711" rel="nofollow" target="_blank">Deep Learning 之 最优化方法</a>系列文章的Adam方法。主要参考Deep Learning 一书。</p>
整个优化系列文章列表:
Deep Learning 之 最优化方法
Deep Learning 最优化方法之SGD
Deep Learning 最优化方法之Momentum(动量)
Deep Learning 最优化方法之Nesterov(牛顿动量)
Deep Learning 最优化方法之AdaGrad
Deep Learning 最优化方法之RMSProp
Deep Learning 最优化方法之Adam
先上结论:
1.Adam算法可以看做是修正后的Momentum+RMSProp算法
2.动量直接并入梯度一阶矩估计中(指数加权)
3.Adam通常被认为对超参数的选择相当鲁棒
4.学习率建议为0.001
再看算法:其实就是Momentum+RMSProp的结合,然后再修正其偏差。
<script>
(function(){
function setArticleH(btnReadmore,posi){
var winH = $(window).height();
var articleBox = $("div.article_content");
var artH = articleBox.height();
if(artH > winH*posi){
articleBox.css({
'height':winH*posi+'px',
'overflow':'hidden'
})
btnReadmore.click(function(){
if(typeof window.localStorage === "object" && typeof window.csdn.anonymousUserLimit === "object"){
if(!window.csdn.anonymousUserLimit.judgment()){
window.csdn.anonymousUserLimit.Jumplogin();
return false;
}else if(!currentUserName){
window.csdn.anonymousUserLimit.updata();
}
}
articleBox.removeAttr("style");
$(this).parent().remove();
})
}else{
btnReadmore.parent().remove();
}
}
var btnReadmore = $("#btn-readmore");
if(btnReadmore.length>0){
if(currentUserName){
setArticleH(btnReadmore,3);
}else{
setArticleH(btnReadmore,1.2);
}
}
})()
</script>
</article>