gabriele renzi

@riffraff
89 Followers
186 Following
2K Posts
Code Monkey, Professional Student, Geek
twitter@riffraff
bloghttps://riffraff.info
blueskyhttps://bsky.app/profile/riffraff.info

wanna know one of the dirty little secrets of how to make a cool science fiction set? this bad boy right here, the NFS 4 Freezer Spacer. this fucker shows up EVERYWHERE in scifi set design

ceilings? check! floors? check! walls? check! tables? check!

instant scifi vibe

One of the numerical exponents that was improved via Deepmind's AlphaEvolve (AE) tool (as part of a collaboration I am involved in) has just been improved: https://arxiv.org/abs/2505.16105

The question concerns the best exponent θ for which one can construct large sets of integers A,B with |A+B| = O(|A|) and |A-B| >> |A|^θ. The previous best construction, due to Gyamarti, Hennecart, and Ruzsa in 2007, gave a lower bound of 1.14465; our group was able to use AlphaEvolve to obtain the improvement 1.1584; and this new paper obtains 1.173050. (The best known upper bound is 4/3.)

All constructions proceed by first locating an auxiliary finite set U of integers with good properties. The original construction of Gyamarti et al. was a set of cardinality 30000 or so, constructed according to a specific recipe; AE did an unconstrained search and found an example of cardinality 54265 that gave the better bound; and the new paper returns to (a modification) of the original construction but with much larger parameters to find (with computer assistance) a set with more than 10^43546 elements.

It is tempting to simplify this to a zero-sum narrative of "winners" and "losers", but I think it is great that different approaches can complement each other here to make mathematical progress. The advantages of an AE-type approach are more in the direction of breadth rather than depth; one can use AE to scan large ranges of problems to identify places where the literature could be improved, and human experts (perhaps also assisted by computers) can then focus attention on these problems to make further progress.

Sums and differences of sets (improvement over AlphaEvolve)

On May 14, 2025, DeepMind announced that AlphaEvolve, a large language model applied to a set of mathematical problems, had matched or exceeded the best known bounds on several problems. In the case of the sum and difference of sets problem, AlphaEvolve, using a set of $54265$ integers, improved the known lower bound of $θ=1.14465$ to $θ=1.1584$. In this paper, we present an improved bound $θ=1.173050$ using an explicit construction of a U set that contains more than $10^{43546}$ elements. For fast integer and floating-point arithmetic, we used the (free) GMP library.

arXiv.org
@fskornia @cstross to be fair, there's a chance the US will start a war over Greenland cause Trump.does not understand map projections, no need for ai slop
@thibault @bagder
It's client controlled, in the '00s abusing this was a way to get P2P working from inside firewalled/NAT'ed networks (universities, some ISPs).
Art of the deal…

Hi Mastodon!
IEEE Spectrum is an award-winning technology magazine, dedicated to bringing you the best technology news and analysis.

https://spectrum.ieee.org/

So after firing thousands of highly trained and conscientious federal scientists, the White House released a health report that was written by AI and full of errors.

https://www.nytimes.com/2025/05/29/well/maha-report-citations.html?unlocked_article_code=1.K08.QwQ1.XQjgiRZggkyJ&smid=nytcore-ios-share&referringSource=articleShare

#health #ai #incompetence #giftarticle #uspol #trump #rfkjr

White House Health Report Included Fake Citations

A report on children’s health released by the Make America Healthy Again Commission referred to scientific papers that did not exist.

The New York Times

I've been building this new tool for awhile, and I ended up with something I am proud of.
Shields.io has never had a full   badger. Now it does. This initial release ships with 6 of my favorite badges, and makes it supremely simple to add more. DSL FTW. I'll add more soon, and I hope you will too.

If you 💓 📛 as much as I do (high information density) then I hope you'll give me a star/follow/toot/skeet.
https://shields-badge.galtzo.com/
#Ruby #RubyGem #Badge

File: README — Documentation by YARD 0.9.37

due to <reasons> I'm working from a cafe-cum-working-tables place and I can't understand people who feel productive in such an environment.

Partly, because I feel like I'm disturbing everyone.

I'm afraid, working from home for 15 years ruined me.

#remote #work #wfh