<?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>Probability on Bdim</title>
    <link>/tags/probability/</link>
    <description>Recent content in Probability on Bdim</description>
    <generator>Hugo -- 0.164.0</generator>
    <language>en</language>
    <copyright>2021-CURRENT_YEAR Bdim</copyright>
    <lastBuildDate>Wed, 24 Dec 2025 00:00:11 +0800</lastBuildDate>
    <atom:link href="/tags/probability/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Learning Notes on HMM (Hidden Markov Model)</title>
      <link>/posts/learning-notes-on-hmm-hidden-markov-model/</link>
      <pubDate>Wed, 24 Dec 2025 00:00:11 +0800</pubDate>
      <guid>/posts/learning-notes-on-hmm-hidden-markov-model/</guid>
      <description>&lt;h1 id=&#34;overview-and-intuitive-understanding&#34;&gt;Overview and Intuitive Understanding&lt;/h1&gt;
&lt;p&gt;Hidden Markov Model (HMM) is a class of &lt;strong&gt;probabilistic models used to model and infer system evolution processes in situations where &amp;ldquo;states are unobservable&amp;rdquo;&lt;/strong&gt;. The core problem it solves can be summarized in one sentence:&lt;/p&gt;
&lt;blockquote&gt;
&lt;p&gt;When the true state of the real world cannot be directly observed, can we still infer the most likely internal state change process of the system based solely on the observed phenomena?&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
