Forwarded from AlexTCH
https://lecopivo.github.io/SciLean/doc/differentiation_in_scilean.html
A cool tutorial (work in progress) on calculating derivatives and differentials (symbolically and mostly automatically) in Lean 4 with the SciLean library (and tactics).
A cool tutorial (work in progress) on calculating derivatives and differentials (symbolically and mostly automatically) in Lean 4 with the SciLean library (and tactics).
👏3👍1
https://arxiv.org/abs/2208.09964 Shor, [2022] The Early Days of Quantum Computation"
🤔1
Forwarded from AlexTCH
https://www.arxiv-vanity.com/
> arXiv Vanity renders academic papers from arXiv as responsive web pages so you don’t have to squint at a PDF.
WOW! 😳
> arXiv Vanity renders academic papers from arXiv as responsive web pages so you don’t have to squint at a PDF.
WOW! 😳
👍5
Forwarded from AlexTCH
https://buttondown.email/hillelwayne/archive/why-you-should-read-data-and-reality/
Эта фраза точечно объясняет, почему ООП по факту провалилось, как и примерно все остальные "методологии программирования" или проектирования.
Реальность многообразнее любой модели — это первое, что мы забываем, и оно же возвращается бумерангом чтобы хлопнуть нас по затылку в самый ответственный момент.
Ссылка на книгу внутри поста.
#free #book #modeling
Once more: we are not modeling reality, but the way information about reality is processed, by people. — Bill Kent
Эта фраза точечно объясняет, почему ООП по факту провалилось, как и примерно все остальные "методологии программирования" или проектирования.
Реальность многообразнее любой модели — это первое, что мы забываем, и оно же возвращается бумерангом чтобы хлопнуть нас по затылку в самый ответственный момент.
Ссылка на книгу внутри поста.
#free #book #modeling
Buttondown
Why You Should Read "Data and Reality"
Once more: we are not modeling reality, but the way information about reality is processed, by people. — Bill Kent I've got this working theory that you can...
👍5👎3🤩1
👍2🤔2
Forwarded from Known unknown unknowns
All physics in nine lines: https://www.motionmountain.net/9lines.html (там ссылки на Motion Mountain, её тоже неплохо почитать)
www.motionmountain.net
All of physics and all of nature in 9 lines
All of physics can be condensed
in 9 short lines. The 9 lines contain all equations and all observations of modern physics.
in 9 short lines. The 9 lines contain all equations and all observations of modern physics.
🤩3👍1
Forwarded from AlexTCH
120. Adapting old programs to fit new machines usually means adapting new machines to behave like old ones.
Alan J. Perlis (http://pu.inf.uni-tuebingen.de/users/klaeren/epigrams.html)
Эта фраза точечно объясняет развитие компьютеров за последние 50 лет.
👍9👏1
Forwarded from Alexander Chichigin
https://www.youtube.com/watch?v=B7aPcZM_JXo
Если кто-то ещё не видел как теорию категорий и (профункторную) оптику применяют к ML (в виде автоматического дифференцирования) прямо в динамически-типизированной Julia. 😏
Если кто-то ещё не видел как теорию категорий и (профункторную) оптику применяют к ML (в виде автоматического дифференцирования) прямо в динамически-типизированной Julia. 😏
YouTube
Keno Fischer: "Optics in the wild: reverse mode automatic differentiation in Julia"
Intercats: 28th June 2022
———
Using categorical inspiration in real world software systems: "I'll definitely be talking about the optics formalism of reverse mode automatic differentiation, but if I have space, I might end up talking about some more recent…
———
Using categorical inspiration in real world software systems: "I'll definitely be talking about the optics formalism of reverse mode automatic differentiation, but if I have space, I might end up talking about some more recent…
🤔3👍2
https://arxiv.org/abs/2209.11339 Faul, Manuell, [2022] "Machine Space I: Weak exponentials and quantification over compact spaces"
arXiv.org
Machine Space I: Weak exponentials and quantification over compact spaces
Topology may be interpreted as the study of verifiability, where opens correspond to semi-decidable properties. In this paper we make a distinction between verifiable properties themselves and...
Forwarded from AlexTCH
https://lemire.me/blog/2019/10/16/benchmarking-is-hard-processors-learn-to-predict-branches/
Dang! CPUs really do learn to predict branches. And learn fast!
If you're trying to benchmark how your code handles "cold" (fresh) data it will screw your tests real good.
Dang! CPUs really do learn to predict branches. And learn fast!
If you're trying to benchmark how your code handles "cold" (fresh) data it will screw your tests real good.
The recordings of ICFP 2022 are available online, https://www.youtube.com/playlist?list=PLyrlk8Xaylp4ee6ZAtFD9XMD2EZ02K9xK
YouTube
ICFP 2022 - YouTube
🤩1
Forwarded from AlexTCH
https://www.deepmind.com/blog/discovering-novel-algorithms-with-alphatensor
If you strip all the nuances DeepMind found a way to represent matrix multiplication as a single-player game with scores proportional to algorithm efficiency and fed it into AlphaZero, which is notoriously good at games. And indeed properly modified AlphaZero dubbed AlphaTensor found new State-of-the-Art matrix multiplication algorithms for a wide range of fixed matrix sizes, including ones optimized for GPGPUs and TPUs specifically.
In a broader context this is indeed a huge leap in applying Reinforcement Learning to algorithms research. Expect a thick stream of papers feeding various kinds of algorithmic problems into more or less the same system.
If you strip all the nuances DeepMind found a way to represent matrix multiplication as a single-player game with scores proportional to algorithm efficiency and fed it into AlphaZero, which is notoriously good at games. And indeed properly modified AlphaZero dubbed AlphaTensor found new State-of-the-Art matrix multiplication algorithms for a wide range of fixed matrix sizes, including ones optimized for GPGPUs and TPUs specifically.
In a broader context this is indeed a huge leap in applying Reinforcement Learning to algorithms research. Expect a thick stream of papers feeding various kinds of algorithmic problems into more or less the same system.
Google DeepMind
Discovering novel algorithms with AlphaTensor
In our paper, published today in Nature, we introduce AlphaTensor, the first artificial intelligence (AI) system for discovering novel, efficient, and provably correct algorithms for fundamental...
🤯2🤔1
In honor of Dana Scott’s 90th birthday,
a Git repository containing pdf scans of
a selection of his papers has been established.
These are available for public download here:
https://github.com/CMU-HoTT/scott
a Git repository containing pdf scans of
a selection of his papers has been established.
These are available for public download here:
https://github.com/CMU-HoTT/scott
GitHub
GitHub - CMU-HoTT/scott: Selected Papers of Dana S. Scott
Selected Papers of Dana S. Scott. Contribute to CMU-HoTT/scott development by creating an account on GitHub.
👍3🎉3
Forwarded from Протестировал (Sergey Bronnikov)
Помните SETI@home? Проект анализировал радиосигнал из космоса для поиска внеземного разума используя для этого вычислительные ресурсы добровольцев. По аналогии с SETI@Home есть проект Fuzzing@Home, в котором можно запускать фаззинг проектов, добавленных в OSS Fuzz, в обычном веб браузере. Это возможно благодаря компиляции в WebAssembly. Попробуйте сами - http://fuzzcoin.gtisc.gatech.edu:8000/
😱3👍2🤩2
👍1