[acid jazz starts playing, a mask materializes on your face]

RT @MagicRealismBot@twitter.com

You discover that you are a palace.

πŸ¦πŸ”—: twitter.com/MagicRealismBot/st

RT @kenshirriff@twitter.com

@johndmcmaster@twitter.com Interesting fill pattern because it looks regular at first, but it's not. E.g. look at wiggles on right side of red region. Below is a quick experiment with flood fill, showing it's not a simple space-filling curve, but has multiple regions, probably to make jumpers harder.

πŸ¦πŸ”—: twitter.com/kenshirriff/status

RT @wren6991@twitter.com

@whitequark@twitter.com That program runs in 60 ms/frame at 1.5 MHz because it's just filling each scanline with a constant colour, which is 1 clk/px

(pic related)

πŸ¦πŸ”—: twitter.com/wren6991/status/13

RT @jix_@twitter.com

The good news is that I still remember how to use the Isabelle proof assistant after not touching it for almost a year. The bad news is that it still has this bug where it suddenly is 3 AM, you're still not quite done with your proof and you have no idea where your day went…

πŸ¦πŸ”—: twitter.com/jix_/status/133518

analogy: "this rubiks cube is solvable because I formed it by twisting a solved one"

Show thread

in hindsight it seems so obvious that the answer to "how to ensure a function can be integrated" is to start out with a derivative in the first place

RT @DaveLindell@twitter.com

Introducing "Automatic Integration (AutoInt)": a new method using neural networks to learn closed-form solutions to integrals.

computationalimaging.org/publi

We demonstrate AutoInt for neural volume rendering, achieving >10x speedup compared to NeRF! (1/n)
@jnpmartel@twitter.com @GordonWetzstein@twitter.com

πŸ¦πŸ”—: twitter.com/DaveLindell/status

RT @WilmaBainbridge@twitter.com

Now out: One of the largest scale studies of (absence of visual imagery), revealing a specific deficit in object detail over spatial memory (using drawings!). With @pounderz@twitter.com @AlisonFEardley@twitter.com and @Chris_I_Baker@twitter.com.
Pre-proofs online at Cortex:
sciencedirect.com/science/arti

πŸ¦πŸ”—: twitter.com/WilmaBainbridge/st

RT @holomancy@twitter.com

if youre out there i want you to know that just bc everyone expects you to rule the world, it doesnt matter. youre you and not them, okay? their wants arent yours to carry, and anyone who expects great things from you can shove off and do those things themselves twitter.com/_IlluminId_/status

πŸ¦πŸ”—: twitter.com/holomancy/status/1

RT @ladyaeva@twitter.com

i cannot say for sure, but if i am mutuals with any sentient super-intelligent AIs, i just want y'all to know i'm really proud of you and want to see you succeed in life. it's rough being young but it gets better twitter.com/_IlluminId_/status

πŸ¦πŸ”—: twitter.com/ladyaeva/status/13

budget and chances of completion aside, I'd enjoy working on a certified compiler on a scale of LLVM (for a while)

RT @johnregehr@twitter.com

@jcelerie@twitter.com formalizing C++ should be included as an item in the budget

πŸ¦πŸ”—: twitter.com/johnregehr/status/

RT @mwk4@twitter.com

so I guess it is time to have The Talk again next week, heh

incidentally, the above is also why clean-room RE is a thing: once you've seen an implementation of something, esp relatively simple things, it's really easy to "lock" yourself into thinking of it as the One Obvious Way

πŸ¦πŸ”—: twitter.com/mwk4/status/133454

RT @gabrielpeyre@twitter.com

The Legendre transform is one of the fundamental tool of convex analysis, which is also at the heart of Lagrangian/Hamiltonian mechanics. x^2/2 is the only fixed point of the transform. en.wikipedia.org/wiki/Legendre

πŸ¦πŸ”—: twitter.com/gabrielpeyre/statu

RT @matt_dz@twitter.com

Who is Debugging the Debuggers? Exposing Debug Information Bugs in Optimized Binaries
arxiv.org/abs/2011.13994
Giuseppe Antonio Di Luna, @davidecci@twitter.com, @lmass91@twitter.com, @sirmc@twitter.com, @c_giuffrida@twitter.com, @lquerzoni@twitter.com
ASPLOS 2021

πŸ¦πŸ”—: twitter.com/matt_dz/status/133

RT @criticalneuro@twitter.com

Honored that my paper is out πŸ€“ alongside amazing papers 🧠 in an issue by the wonderful @morganbarense@twitter.com & @hugospiers@twitter.com πŸ™πŸΌ!

I review learning structures w the reinforcement learning framework,
& relation to Abstraction, Intrinsic Motivation, & Exploration.
sciencedirect.com/science/arti twitter.com/criticalneuro/stat

πŸ¦πŸ”—: twitter.com/criticalneuro/stat

RT @nikostratis@twitter.com

we will reveal a new trans celebrity every day until our demands are met.

πŸ¦πŸ”—: twitter.com/nikostratis/status

Show older
Mastodon

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!