<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Cross-Entropy on Bdim</title>
    <link>/tags/cross-entropy/</link>
    <description>Recent content in Cross-Entropy on Bdim</description>
    <generator>Hugo -- 0.164.0</generator>
    <language>en</language>
    <copyright>2021-CURRENT_YEAR Bdim</copyright>
    <lastBuildDate>Wed, 18 Sep 2024 00:01:11 +0800</lastBuildDate>
    <atom:link href="/tags/cross-entropy/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Understanding Cross-Entropy</title>
      <link>/posts/understanding-cross-entropy/</link>
      <pubDate>Wed, 18 Sep 2024 00:01:11 +0800</pubDate>
      <guid>/posts/understanding-cross-entropy/</guid>
      <description>&lt;h2 id=&#34;what-is-it&#34;&gt;What is it?&lt;/h2&gt;
&lt;p&gt;In deep learning, &amp;ldquo;cross-entropy&amp;rdquo; is a function used to measure the difference between two probability distributions, typically used in classification tasks.&lt;/p&gt;
&lt;h3 id=&#34;analogy-guessing-the-color-of-candies&#34;&gt;Analogy: Guessing the Color of Candies&lt;/h3&gt;
&lt;p&gt;Imagine you have a bag of candies containing red, green, and blue colors. You guess the color distribution of the candies is 50% red, 30% green, and 20% blue. However, the actual distribution is 60% red, 20% green, and 20% blue.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
