<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Bayes on Alok Regmi</title>
		<link>https://alokregmi.com/tags/bayes/</link>
		<description>Recent content in Bayes on Alok Regmi</description>
		<generator>Hugo</generator>
		<language>en-us</language>
		
		
		
		
			<lastBuildDate>Wed, 12 Aug 2020 13:07:15 +0545</lastBuildDate>
		
			<atom:link href="https://alokregmi.com/tags/bayes/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Bayesian Prior</title>
				<link>https://alokregmi.com/braindump/2020/04/bayesian-prior/</link>
				<pubDate>Sun, 19 Apr 2020 15:58:50 +0545</pubDate>
				<guid>https://alokregmi.com/braindump/2020/04/bayesian-prior/</guid>
				<description>&lt;p&gt;&amp;ndash; Always set the priors in the vicinity of what you believe the truth is&#xA;&amp;ndash; Always set the priors such that they reflect the same order of magnitude&#xA;as the phenomenon you’re trying to predict&#xA;&amp;ndash; Don’t be overconfident, leave space for doubt&#xA;&amp;ndash; Never use completely uninformative priors&#xA;&amp;ndash; Whenever possible refrain from using uniform distributions&#xA;&amp;ndash; Always sum up the consequence of all of your priors such&#xA;that if no data was available your model still predicts&#xA;in the same order of magnitude as your observed response&#xA;&amp;ndash; Be careful, and be honest! Never postulate very informative priors on&#xA;results you WANT to be true. It’s OK&#xA;if you BELIEVE them to be true.&#xA;Don’t rest your mind until you see the difference.&lt;/p&gt;</description>
			</item>
			<item>
				<title>Bayesian Prior</title>
				<link>https://alokregmi.com/ai/bayesianprior/</link>
				<pubDate>Mon, 20 Jan 2020 00:00:00 +0000</pubDate>
				<guid>https://alokregmi.com/ai/bayesianprior/</guid>
				<description>&lt;ul&gt;&#xA;&lt;li&gt;Always set the priors in the vicinity of what you believe the truth is&lt;/li&gt;&#xA;&lt;li&gt;Always set the priors such that they reflect the same order of magnitude&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&lt;p&gt;s the phenomenon you’re trying to predict&lt;/p&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;Don’t be overconfident, leave space for doubt&lt;/li&gt;&#xA;&lt;li&gt;Never use completely uninformative priors&lt;/li&gt;&#xA;&lt;li&gt;Whenever possible refrain from using uniform distributions&lt;/li&gt;&#xA;&lt;li&gt;Always sum up the consequence of all of your priors such&#xA;that if no data was available your model still predicts&#xA;in the same order of magnitude as your observed response&lt;/li&gt;&#xA;&lt;li&gt;Be careful, and be honest! Never postulate very informative priors on&#xA;results you WANT to be true. It’s OK&#xA;if you BELIEVE them to be true.&#xA;Don’t rest your mind until you see the difference.&lt;/li&gt;&#xA;&lt;/ul&gt;</description>
			</item>
	</channel>
</rss>
