In other news. OpenBSD project released a new version control system implemented on top of (and thus fully compatible with) Git repository format: https://gameoftrees.org/

I for one have no idea who needs this apart from OpenBSD developers. Windows naturally is not supported (though MacOS is).

https://theconversation.com/shifting-ocean-currents-are-pushing-more-and-more-heat-into-the-southern-hemispheres-cooler-waters-189122

Everyone knows fluid dynamics is a bitch. On the global scale it becomes even funnier. But not once shit will hit the fan for reals...

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.

#madeuppaper
"Horror in modern video games: state of the frightening art"

https://nextjournal.com/

Hosted "polyglot" (whatever you put in a Docker) Notebooks in the same #notebook. Supports GPU even on the free plan.

https://github.com/nextjournal/clerk

#notebook style Smalltalk-inspired interactive environment for Clojure connecting to any text editor.

https://avehtari.github.io/ROS-Examples/

A deep #book on statistics, regression (of various kinds) and causal inference from some of the best researchers and practicioners. With a #free PDF version available (and source code of examples of course).

Я бы, конечно, предпочёл __годы__ здорового сердца, но чёрт с ним, уже согласен и на дни... 😏

https://www.youtube.com/watch?v=c33AZBnRHks

Офигенно весёлая история задротской оптимизации никому не нужного куска кода — ровно то, что мы любим больше всего! 😂

Возлюби ближнего как кота своего!

https://statmodeling.stat.columbia.edu/2020/07/02/no-i-dont-believe-that-claim-based-on-regression-discontinuity-analysis-that/

A great post with thorough replication/reanalysis/discussion and we might say debunking. Also an example of pretty decent scientific discussion. Plus deep technical dives in the comments.

#statistics #rdd

На днях летел в Airbus A319 — приятно сознавать, что софт, им управляющий, был верифицирован (в разных отношениях) и собран CompCert. 😏

Нашёл в темнушке старый обоссаный котами бокен.
Атака +20
Привлекательность -50

Когда в 2005 читал "Метро 2033" казалось, что автор установил дату ядерного апокалипсиса слишком близко... 😏

А вообще, про "Метро" вспомнил в связи с "локативным искусством" (привязанным к местности) — пока не покатался на московском метро, роман как-то не резонировал примерно никак.

https://www.youtube.com/watch?v=CuxfRdjfZG0
Oldie but goodie. Curious highlights.

Climate models have surprisingly low error rates in terms of software bugs. It's nice that people really can and do produce high-quality software without Sales and Marketing pressure. 😏

The models systematically overestimate some parameters but underestimate others. Thus when you see graphs projecting climate into the far future, actual air temperatures are likely to be a bit lower (while deep ocean temperatures will be higher). And the see levels are likely to be higher than predicted which is not very nice.

Make sense not war

https://github.com/stan-dev/stanc3

The new(ish, compared to the previous one) (source-to-source) compiler for the Stan language is implemented in OCaml and became the default one quite some time ago. The Readme also has a reference to a discussion "OCaml vs. Rust". Mind that the discussion take place back in 2018 (4 years ago! 😱) and you can appreciate how much Rust has matured during the time.

And Stan is a huge (by impact) project in the area of Bayesian modeling, statistical analysis and inference.

And speaking about statistics...

https://statmodeling.stat.columbia.edu/2022/11/08/dying-children-and-post-publication-review-rectal-suppositories-edition/

Sloppy statistical analysis indeed jeopardizes lives of actual children in Africa. 😞

Attempting to prove any nontrivial theorem about your program will expose lots of bugs: The particular choice of theorem makes little difference!

LOL, so true! 😂
From https://statmodeling.stat.columbia.edu/2022/11/09/simulation-based-calibration-checking-sbc-is-stronger-than-you-thought-and-the-sbc-package-in-r/#comment-2122385

Кот — не карась, не харрась кота!