How to get URL link on X (Twitter) App
https://twitter.com/ylecun/status/1655609954445717506But I have substantial hope about making this work:
https://twitter.com/AutismCapital/status/1592406121368543233related: SBF was indeed wrong about the St Petersburg paradox
https://twitter.com/TaylorPearsonMe/status/1590826638429650944?s=20&t=i6NYFAMvKddcyXbd6wiHzg
https://twitter.com/KerryLVaughan/status/1545060355432337409for the record, I don't endorse Kerry's thread; the vitriol level is too high for my taste
https://twitter.com/juan_cambeiro/status/1545121334849683456
https://twitter.com/jjcarett2/status/1543948734362537991(to be clear checking category-theoretic ideas with the Zulip is actually a very reasonable and good move. it's hilarious how much expertise in this area has become orthogonal to academic hierarchies and that's not a bad thing!)
https://twitter.com/robbensinger/status/1543532796815126530@CineraVerinia expressed strong suspicion of this slingshotting, and I think rightly so. In practice, even in the absence of deliberate contrarianism for its own sake, there will be a kind of winner's curse among expert opinions. I labeled this as case 4.
https://twitter.com/CineraVerinia/status/1543536202468450304
https://twitter.com/kareem_carr/status/1293942267410108417ℝ is constructed step-by-step:
https://twitter.com/sir_deenicus/status/1292173738541285378…because there are no other operations that return this arbitrary type R. Meanwhile Dist(X) is more like ∀(R:Convex).(X→R)→R. Now I can give the continuation a bunch of values in X and weighted-average the return values. So a Dist(X) is a bunch of X’s associated with weights.
https://twitter.com/_onionesque/status/1210449910467874816With the invertibility assumption, this *is* a manifold, but an extremely special type of manifold: while manifold theory offers a lot of complexity around gluing together local Euclidean "charts", typical representation-learning learns just one *global* Euclidean chart! (2/n)