The Implicit Bias of Gradient Descent on Separable Data · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/1710.10345/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTThe Implicit Bias of Gradient Descent on Separable Data — mentioned by math_dandyhttps://news.ycombinator.com/item?id=44946215urn:hackerpapers:mention:1710.10345:44946215Mon, 18 Aug 2025 22:58:36 GMT<p>I don&#x27;t buy the narrative that the article is promoting. I think the machine learning community was largely over overfitophobia by 2019 and people were routinely using overparametrized models capable of interpolating their training data while still generalizing well. The Belkin et al. paper wasn&#x27;t heresy. The authors were making a technical point - that certain theories of generalization are incompatible with this interpolation phenomenon. The lottery ticket hypothesis paper&#x27;s demonstration of the ubiquity of &quot;winning tickets&quot; - sparse parameter configurations that generalize - is striking…</p><p><a href="https://hackerpaper.kaipfister.de/p/1710.10345/">Paper and discussion history</a>. A mention is not an endorsement.</p>stat.ML