krnsll

8 ms·

1.
by krnsll 1y ago
If you think of this as a search, retrieval and “application” problem on the space of convex optimization proof techniques, it’s not a particularly striking result to a mathematician. Partly because: the space of results/techniques and
2.
by krnsll 1y ago
Agreed. Something else that might be driving this is that existing models essentially get the job done for most users, who —- unlike HN commenters (I promise this is human generated, em dash notwithstanding ;P) —- don't quite about the
3.
by krnsll 1y ago
Sure Klartag isnt a sphere packing specialist by training, but he's one of the best problem solvers around. He just resolved the Hyperplane Conjecture earlier this year and has contributed to progress on related problems in convexity t
4.
by krnsll 1y ago
Not sure I agree. It does a decent job of conveying the essential idea for a broader readership: perturb a graph through its adjacency matrix just enough to make the universality conjecture hold for the distribution of eigenvalues -> ana
5.
by krnsll 1y ago
> I wonder how helpful it would be for music and art education if AI could help us deconstruct some songs. Was thinking along the same lines as I fell into the trap of Ghiblifying pictures earlier this week. As someone who spent countles
6.
by krnsll 2y ago
As a mathematician, I can't help but simmer each time I find the profession's insistence on grasping the how's and why's of matters to be dismissed as pedantry. Actionable results are important but absent understanding,
7.
by krnsll 2y ago
Not at all, the proof was a very elegant argument involving fourier transforms and an integral estimate going back to the study of controlling random walks (Khintchine's inequality). I say elegant in the manner of it being enviably so
8.
by krnsll 2y ago
You're on the right track. My comment alludes to sections, but there's a lot of essentially analogous math that explains the phenomena via shadows/"projections."
9.
by krnsll 2y ago
Something I worked on in my PhD was analyzing high dimensional bodies via their "sections." Here's the Busemann Petty Problem: Given two origin symmetric convex bodies K and L in n dimensions. Suppose for every linear hyperpl
10.
by krnsll 4y ago
If you're willing to pay, Interintellect might be it. https://interintellect.com/
11.
by krnsll 4y ago
Previous discussion from 2018: https://news.ycombinator.com/item?id=17873284
12.
by krnsll 5y ago
Ah yes, learning (imitating?) how not to imitate from a listicle that itself cites advice from a compendium of books and sources. Sorry, couldn't help myself.
13.
by krnsll 5y ago
I don't think software developers/professionals have acquired the cultural capital to be viewed as elites quite yet. I think much of the antagonistic Silicon Valley coverage in publications like NYT, which one could say are more t
14.
by krnsll 5y ago
Great recommendation. When I first listened to this podcast, I was impressed by how the conversation managed to hit the sweet spot of engagement for both non technical (a friend I shared it with) and technical/mathematical (myself) aud
15.
by krnsll 6y ago
Thanks for pointing that out. My apologies for missing Tomonaga. Was going on my anecdotal memory, should have researched.
16.
by krnsll 6y ago
Freeman Dyson is a glaring miss here. Feynman and Schwinger shared the 1965 Nobel in Physics for their respective perspectives on Quantum Electrodynamics. But it was Dyson who unified Feynman's diagram approach with Schwinger's fi
17.
“Economics with a Moral Compass?”: Amartya Sen and Angus Deaton(annualreviews.org)
1 points by krnsll 6y ago|0 comments
18.
by krnsll 6y ago
Seconded. Read Lasch (Revolt of the Elites) to make sense of what dyed-in-wool Marxists think of the Woke "ideology" (if we can even be charitable enough to use that word, for most of what they say lacks coherence). In fact, even
19.
Tech Companies Want You to Believe America Has a Skills Gap(bloomberg.com)
7 points by krnsll 6y ago|0 comments
20.
by krnsll 6y ago
1) These two single volume histories were good starts for me: a) John Keay's "China: A History" and b) Fairbank/Goldman's "China: A New History". 2) HUP's History of Imperial China six volume series e
21.
by krnsll 6y ago
The point about your mom (and more broadly older faculty) is important. I hope it works out for her. Vulnerable faculty will likely be forced to hold "in-person" lectures (ergo worsening their risk profiles) for the universities t
22.
by krnsll 6y ago
John Cochrane, over at Hoover wrote about university finances in light of Covid19 a few months back. Worth a read. https://johnhcochrane.blogspot.com/2020/04/university-financ...
23.
Larry Summers: How to Fix Globalization–For Detroit, Not Davos(the-american-interest.com)
1 points by krnsll 6y ago|0 comments
24.
John Cochrane on University Finances(johnhcochrane.blogspot.com)
2 points by krnsll 6y ago|1 comments
25.
by krnsll 6y ago
I've always found the mutual knowledge vs common knowledge dichotomy to be a good way to conceptualize this distinction between kayfabe and propaganda (although it is not strictly speaking the perfect analogy here, it does offer some i
26.
Ben Thompson (Stratechery) and Bill Simmons on the “Streaming Wars” [audio](open.spotify.com)
1 points by krnsll 7y ago|0 comments
27.
by krnsll 7y ago
Would be interested to see what Judea Pearl says about this. Not on twitter any more but back when I was, I recall him regularly tweeting about causality and AI's shortcomings in that regard.
28.
by krnsll 7y ago
Thanks for sharing this. The piece makes an urgent point that needs to be appropriately acknowledged. It falls in the vein of arguments I've come to term as "relational poverty" or "relational inequality" arguments
29.
by krnsll 7y ago
Thanks for sharing this. I'm inclined to agree with your observations/reflections. Moreover, I suspect this might be common to the post-colonial experience [fn0] , if we account for: (a) what life was like under colonial rule; (b
30.
by krnsll 7y ago
An afterthought: I suspect this is done subconsciously too quite often by people in such fields as part of their evidence seeking (hypothesis confirmation) procedures which are admittedly more adhoc than those sort of naturally built into m

More ›