https://hackerpaper.kaipfister.de/p/1906.05849/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=26696701urn:hackerpapers:mention:1906.05849:26696701Mon, 05 Apr 2021 08:49:56 GMT<p>I will nitpick and say that models don&#x27;t compress data but extract useful information. Compression lossly or lossless is for data reconstruction. Machine learning models retain information that generalises https://arxiv.org/abs/1906.05849</p><p><a href="https://hackerpaper.kaipfister.de/p/1906.05849/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CV