Alexey Stysin retweeted
every 10 like-o’s this get i use BI (bird intelligenkcse) (chirpBRD) to make this drawing more art
Alexey Stysin retweeted
this is the ideal mathematician body.
you may not like it but this is what peak performance looks like
Construction of GPT-5
Pieter Bruegel the Elder
Oil on wood panel
C. 1563
@progpaintings
53
Alexey Stysin retweeted
Announcement: we're going to be launching ARC 2, a larger & more diverse successor to the ARC challenge -- an exciting new benchmark for AI reasoning abilities.
You're invited to take part in creating it: ARC 2 content will be crowdsourced.
Sign up now: arc-editor.lab42.global/
The Leetcode Triptych
I. Easy II. Medium III. Hard
Hieronymus Bosch 1490-1510
Oil painting on oak panel
Museo del Prado, Madrid
@progpaintings @LeetCode
1
Alexey Stysin retweeted
More fun publisher surveillance:
Elsevier embeds a hash in the PDF metadata that is *unique for each time a PDF is downloaded*, this is a diff between metadata from two of the same paper. Combined with access timestamps, they can uniquely identify the source of any shared PDFs.
Alexey Stysin retweeted
My first impressions of web3
“What I found most interesting, though, is that after OpenSea removed my NFT, it also 𝗻𝗼 𝗹𝗼𝗻𝗴𝗲𝗿 𝗮𝗽𝗽𝗲𝗮𝗿𝗲𝗱 𝗶𝗻 𝗮𝗻𝘆 𝗰𝗿𝘆𝗽𝘁𝗼 𝘄𝗮𝗹𝗹𝗲𝘁 𝗼𝗻 𝗺𝘆 𝗱𝗲𝘃𝗶𝗰𝗲. This is web3, though, how is that possible?”moxie.org/2022/01/07/web3-fi…
Alexey Stysin retweeted
This is Maggie. Playing in the snow was her favorite thing in the world, and she got to do it one final time thanks to @SLCOParksandRec. Their staff shaved extra ice, loaded it into bins, and gave it to her family, making Maggie’s last day very very special. 14/10 for all ❤️
Alexey Stysin retweeted
Replying to @nick_kapur
in English (borrowed from the French) we would call this a round-robin. bartleby.com/81/14554.html
What every computer scientist should know about floating-point arithmetic
Papyrus manuscript, c. 1250 BCE
@progpaintings
Alexey Stysin retweeted
Hi #ml #optimization twitter!
Recently @fpedregosa and I have been working on a *constrained* optimization library for PyTorch. github.com/openopt/chop
Why constrained? Scale matters! Adversarial examples are one nice example of this.
Alexey Stysin retweeted
#つぶやきGLSL
vec2 p;float i,g,d=1.;for(;++i<128.&&d>.001;g+=d=-(length(p)-2.+g/9.)/2.)p=vec2((FC.xy-.5*r)/r.y)*g+vec2(.3)*rotate2D(g*2.);p=vec2(atan(p.x,p.y),g)*8.28+t*2.;p=abs(fract(p+vec2(0,.5*ceil(p.x)))-.5);o+=30./i-.5/step(.9,1.-abs(max(p.x*1.5+p.y,p.y*2.)-1.));






























