https://hackerpaper.kaipfister.de/p/cs_0111054/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=46990738urn:hackerpapers:mention:cs_0111054:46990738Thu, 12 Feb 2026 16:24:13 GMT<p>I&#x27;m doing my PhD in compression-based machine learning, wanted to contribute a few clarifying points. The relationship between probabilistic modeling and lossless compression is very direct. A model that predicts the next symbol with probability p can on average losslessly compress that symbol with the help of an entropy coder (e.g. arithmetic coding) in -log(p) bits. Therefore improved probabilistic models immediately translate into improved lossless compressors. There are two ways people have used compressors for ML: the first is based on the Minimum Description Length (MDL) principle [0]…</p><p><a href="https://hackerpaper.kaipfister.de/p/cs_0111054/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CChttps://news.ycombinator.com/item?id=27047900urn:hackerpapers:mention:cs_0111054:27047900Wed, 05 May 2021 09:32:30 GMT<p>Thanks, I had not seen that dissertation yet, it has much more information than that arxiv link - here is a a pdf of that dissertation that google gave me: https://www.illc.uva.nl/Research/Publications/Dissertations/... Although strictly speaking the idea with the evolutionary trees is a bit older, as far as I know it is from https://arxiv.org/pdf/cs/0111054.pdf</p><p><a href="https://hackerpaper.kaipfister.de/p/cs_0111054/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CC