<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Demo on Luke Salamone&#39;s Blog</title>
		<link>https://blog.lukesalamone.com/tags/demo/</link>
		<description>Recent content in Demo on Luke Salamone&#39;s Blog</description>
		<generator>Hugo</generator>
		<language>en-us</language>
		
		
		
		
			<lastBuildDate>Tue, 16 Sep 2025 17:57:57 -0700</lastBuildDate>
		
			<atom:link href="https://blog.lukesalamone.com/tags/demo/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Can a Neural Net Learn It?</title>
				<link>https://blog.lukesalamone.com/posts/can-a-neural-net-learn/</link>
				<pubDate>Tue, 16 Sep 2025 17:57:57 -0700</pubDate>
				<guid>https://blog.lukesalamone.com/posts/can-a-neural-net-learn/</guid>
				<description>&lt;p&gt;A neural network is a type of function originally inspired by how the brain works. However, what makes it particularly useful in machine learning are the following properties:&lt;/p&gt;&#xA;&lt;ol&gt;&#xA;&lt;li&gt;Neural networks can approximate any continuous function (&lt;a href=&#34;https://en.wikipedia.org/wiki/Universal_approximation_theorem&#34;&gt;universal approximation theorem&lt;/a&gt;)&lt;/li&gt;&#xA;&lt;li&gt;We have an effective way to fit a neural network to input data, called &lt;a href=&#34;https://en.wikipedia.org/wiki/Backpropagation&#34;&gt;backpropagation&lt;/a&gt;.&lt;/li&gt;&#xA;&lt;li&gt;Using the neural network (forward propagation) and training the network (backpropagation) can both be done very efficiently using existing hardware (GPUs).&lt;/li&gt;&#xA;&lt;/ol&gt;&#xA;&lt;p&gt;Below you can experiment with fitting a small neural net to a function and watch as it learns. You can select sample functions from the dropdown, write your own custom function, and specify the (fully connected) neural network structure.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
