False discovery rate smoothing · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/1411.6144/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTFalse discovery rate smoothing — mentioned by tanseyhttps://news.ycombinator.com/item?id=8899739urn:hackerpapers:mention:1411.6144:8899739Fri, 16 Jan 2015 16:11:19 GMT<p>Indeed! I have a particularly relevant horror story. For one of my graduate classes, I built a game where two people (a liar and a truth teller) would try to convince a third person (the judge) they had experienced something. The goal was to see if people used different language when lying than they did when telling the truth. It was a fun project and we were focused more on the NLP/ML side of things. Turns out to be reasonably possible to separate our ~500 examples via a linear SVM, with some interesting separating words. That&#x27;s where our claims stopped. Then I took the results to a…</p><p><a href="https://hackerpaper.kaipfister.de/p/1411.6144/">Paper and discussion history</a>. A mention is not an endorsement.</p>stat.ME