Universal Neural Functionals · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2402.05232/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTUniversal Neural Functionals — mentioned by tudorwhttps://news.ycombinator.com/item?id=39418177urn:hackerpapers:mention:2402.05232:39418177Sun, 18 Feb 2024 11:22:08 GMT<p>Maybe! I&#x27;m lost in the maths, &quot;Our results suggest that learned optimizers can benefit from considering the (symmetry) structure of the weight space they optimize. &quot; this from 7th Feb came out of DeepMind; https://arxiv.org/abs/2402.05232 When it comes to the math underneath an LLM, https://medium.com/autonomous-agents/part-8-mathematical-exp... is about the most accessible explanation I have found so far.</p><p><a href="https://hackerpaper.kaipfister.de/p/2402.05232/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG