<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Scatterplot of KN/PYP language model results</title>
	<atom:link href="http://brenocon.com/blog/2014/02/scatterplot-of-knpyp-language-model-results/feed/" rel="self" type="application/rss+xml" />
	<link>http://brenocon.com/blog/2014/02/scatterplot-of-knpyp-language-model-results/</link>
	<description>cognition, language, social systems; statistics, visualization, computation</description>
	<lastBuildDate>Tue, 25 Nov 2025 13:11:20 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	
	<item>
		<title>By: Chris</title>
		<link>http://brenocon.com/blog/2014/02/scatterplot-of-knpyp-language-model-results/#comment-1163334</link>
		<dc:creator>Chris</dc:creator>
		<pubDate>Wed, 19 Feb 2014 05:02:33 +0000</pubDate>
		<guid isPermaLink="false">http://brenocon.com/blog/?p=2283#comment-1163334</guid>
		<description><![CDATA[One thing I&#039;ve noticed about these HPYP LMs is that when you fit the discount parameter, it is quite small for lower order n-grams, so it is well approximated by a DP (and hence, a good justification for why the DPLM does well on the bigram case).]]></description>
		<content:encoded><![CDATA[<p>One thing I&#8217;ve noticed about these HPYP LMs is that when you fit the discount parameter, it is quite small for lower order n-grams, so it is well approximated by a DP (and hence, a good justification for why the DPLM does well on the bigram case).</p>
]]></content:encoded>
	</item>
</channel>
</rss>
