<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Transfer learning for time series classification · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1811.01533/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/1811.01533/feed.xml" rel="self" type="application/rss+xml" /><item><title>Transfer learning for time series classification — mentioned by mlucy</title><link>https://news.ycombinator.com/item?id=19164826</link><guid isPermaLink="false">urn:hackerpapers:mention:1811.01533:19164826</guid><pubDate>Thu, 14 Feb 2019 19:06:45 GMT</pubDate><description>&lt;p&gt;I don&amp;#x27;t work with time series data much myself. I would imagine you can get at least some transfer learning, since there are patterns that show up across different domains. It looks like there&amp;#x27;s been a little bit of work done on this: https://arxiv.org/pdf/1811.01533.pdf . According to them, transfer learning can improve a time series model if you pick the right dataset to transfer from, but they don&amp;#x27;t seem to be getting the same unbelievably strong transfer results that you&amp;#x27;d see on images and text.&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1811.01533/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item></channel></rss>
