CIFAR10 to Compare Visual Recognition Performance between Deep Neural Networks and Humans · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/1811.07270/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTCIFAR10 to Compare Visual Recognition Performance between Deep Neural Networks and Humans — mentioned by tysam_andhttps://news.ycombinator.com/item?id=34577896urn:hackerpapers:mention:1811.07270:34577896Mon, 30 Jan 2023 10:12:29 GMT<p>93.91%+-1.52! Source: https://arxiv.org/abs/1811.07270 (Table 1 or so I think? :D :)))) Upper limit with (I think very large scale?) training on different data, then transferring to CIFAR10 I believe is 99.5%, but that is extremely unreasonable for commodity algorithms and hardware, unfortunately, I believe D: Hope this helps, thank you for your comment, I really appreciate it! :D</p><p><a href="https://hackerpaper.kaipfister.de/p/1811.07270/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CV