<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>DRL on Arshad Siddiqui</title><link>https://arshadhs.github.io/tags/drl/</link><description>Recent content in DRL on Arshad Siddiqui</description><generator>Hugo</generator><language>en-us</language><atom:link href="https://arshadhs.github.io/tags/drl/index.xml" rel="self" type="application/rss+xml"/><item><title>Reinforcement Learning</title><link>https://arshadhs.github.io/docs/ai/045-deep-reinforcement-learning/010-intro-to-reinforcement-learning/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://arshadhs.github.io/docs/ai/045-deep-reinforcement-learning/010-intro-to-reinforcement-learning/</guid><description>&lt;h1 id="reinforcement-learning">
 Reinforcement Learning
 
 &lt;a class="anchor" href="#reinforcement-learning">#&lt;/a>
 
&lt;/h1>
&lt;p>Reinforcement Learning (RL) is a learning approach in which an &lt;strong>agent&lt;/strong> improves its behaviour by interacting with an &lt;strong>environment&lt;/strong> and observing the rewards produced by its actions.&lt;/p>
&lt;p>Unlike supervised learning, the agent is not given a correct action label for every situation. It must discover useful behaviour through trial, feedback, and repeated interaction.&lt;/p>
&lt;span style="color: blue;">
 &lt;strong>Reinforcement learning is goal-oriented learning through interaction.&lt;/strong>
&lt;/span>
&lt;hr>
&lt;h2 id="why-reinforcement-learning">
 Why Reinforcement Learning?
 
 &lt;a class="anchor" href="#why-reinforcement-learning">#&lt;/a>
 
&lt;/h2>
&lt;p>RL is useful when decisions affect what happens next and the quality of an action may only become clear later.&lt;/p></description></item></channel></rss>