https://hackerpaper.kaipfister.de/p/2006.07464/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=26027911urn:hackerpapers:mention:2006.07464:26027911Thu, 04 Feb 2021 17:56:53 GMT<p>Hey Mike - I did some work on an industry active learning system a few years ago. The high level finding was that transfer &amp; nonparametric methods were huge wins, but online uncertainty and &quot;real active learning&quot; hardly worked at all and were super complicated (at least in caffe1 anyway lol). Can you point to any big breakthroughs that have helped in recent years? Linear Hypermodels (https://arxiv.org/abs/2006.07464) seem promising, but that original experience has left me with some healthy skepticism.</p><p><a href="https://hackerpaper.kaipfister.de/p/2006.07464/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG