https://dmkpress.com/catalog/computer/statistics/978-5-93700-349-2/

Это ДМК перевёл на русский язык
https://matheusfacure.github.io/python-causality-handbook/landing-page.html

Лично я — всецело и полностью одобряю. Во-первых — популяризация #causalinference среди русскоязычной аудитории (кто ещё не освоил английский). Во-вторых — возможность полистать бумажную книгу. Ну и в-третьих — материал-то отличный!

Для тех, кто ещё не перешёл по ссылке, несколько "highlights".

Книга покрывает очень широкий спектр вопросов и методов "причинного вывода" начиная с Randomized Controlled Trials в качестве "золотого стандарта" и далее по Graphical Causal Models, Instrumental Variables, Synthetic Control, Difference-in-Differences, Propensity Scores и всё остальное.

Довольно нетипично, что вторая часть рассказывает про некоторые "EconML" казуальные модели — Meta-Learners, Double ML, вот это вот всё.

Наличествует огромное количество объяснений и примеров на Python, математические формулы (и некоторые теоремы) тоже присутствуют.

https://vezwork.github.io/drostes-lair-post/

A cute cool project. If you don't understand recursion, this can help immensely. Also if you don't know what the amb is, you can and should learn.

If you already know all that, it's just a fun game. 😊

https://www.youtube.com/watch?v=97PbIZBcL6c

So glad I didn't miss this one!

Any nation that draws too great a distinction between its scholars and its warriors will have its thinking done by cowards and its fighting done by fools.

Thucydides as I was told. Curious point. 😏

https://www.broadinstitute.org/news/gene-editing-extends-lifespan-mouse-model-prion-disease

Roughly speaking, they developed a CRISPR-Cas9-based "base editor" that "installs" R37X mutation into a DNA and a pair of adeno-associated viruses (AAVs) to do it in-vivo. R37X mutation messes up prion production in cells which itself are messed up toxic forms of proteins. Lethal forms of proteins.

So this is hugely important. A big step towards treating a disease lethal and uncurable at the moment.

https://www.inkandswitch.com/ambsheets/notebook/01/

Another — surprisingly practical — application of the amb operation. 😊

TIL: turns out there's a notion of "plural naturals", i.e. natural numbers greater than 1.

https://sites.google.com/view/rustverify2025

The Fifth Rust Verification Workshop is coming. To Canada.

https://siek.blogspot.com/2012/07/crash-course-on-notation-in-programming.html

Apparently, Jeremy Siek wrote a very brief "Crash Course on Notation in Programming Language Theory" more than 10 years ago.

https://calculatingempires.net/

A fascinating visualization (a timeline) of the development of communication, data management, computing, education, medicine, economy, energy, policy, surveillance, military and so on from 1500 till today across the world.

I'm not sure it's 100% historically accurate, but still illuminating.

https://www.youtube.com/watch?v=CVLSA8YGvM8

https://catcolab.org (https://github.com/ToposInstitute/CatColab)

This is mind-blowing! 🤯

A collaborative (as in Google Docs) notebook for defining and simulating a range of "logic"/diagram notations starting with purely descriptive ontologies and up to quantitative stock-and-flow diagrams (which define a system of differential equations).

Implemented in Rust and TypeScript offloading the heavy-lifting of actual processing to AlgebraicJulia and Julia's differential equations ecosystem. All based on Double Category Theory.

Gordon Plotkin writes that Rod Burstall recently passed away. 😞

As you can see from the Wikipedia page, among great many other things Burstall was a doctoral advisor for Thorsten Altenkirch, Mike Gordon, Conor McBride, Plotkin himself and other well-known researchers.

https://distrinet.cs.kuleuven.be/jobs/jobs

Do you love building rock-solid code with uncompromising correctness? Do you enjoy hacking in Agda or Coq for hours, squeezing the most out of (dependent) types, and pushing proof assistants to their limits? Does it give you a warm, fuzzy feeling to know that what you are building comes with high-assurance machine-checkable guarantees? If so, we have the perfect PhD position for you!

We offer a fully funded PhD position, including a travel budget, under the supervision of Dominique Devriese and Steven Keuchel in the DistriNet research group of the Department of Computer Science at KU Leuven, Belgium. You would work on scaling up our efforts in mechanizing the verification of security properties in critical low-level code.

Sounds sexy. 😏

Closing date for applications: March 16, 2025.

https://www.andreipopescu.uk/MGS_Sheffield/MGS2025.html
Midlands Graduate School in the Foundations of Computing Science 2025
University of Sheffield, 7-11 April 2025

https://spli.scot/splv/2025-edinburgh/
Scottish Programming Languages and Verification Summer School 2025
University of Edinburgh - 21st to 25th July

https://www.codingfont.com/

Pick your favourite coding font by a tournament between them! 😁

Funny enough, for me won the font I actually use everywhere.

https://arxiv.org/abs/2501.00663
Titans: Learning to Memorize at Test Time
Ali Behrouz, Peilin Zhong, Vahab Mirrokni

One of the last papers of the last year. 😏

Didn't read and most likely wouldn't, but the abstract sounds curious.

Another BOB conference is coming March, 14 2025:
https://bobkonf.de/2025/en/program.html

Among the speakers are:
— Annette Bieniusa
— Andres Löh
— Gillaume Allais
— Manuel Chakravarty
and many others.

TIL: https://logic.math.su.se/seminar/

From there I've learned of https://sinhp.github.io/groupoid_model_in_lean4/

And there are other curious links...

https://www.pm.inf.ethz.ch/research/verifythis.html

Another VerifyThis Competition is coming May 3rd and 4th, 2025.

Online participants are welcome: https://docs.google.com/forms/d/e/1FAIpQLScOhK5tMG5q5DMb36yrTQ8itUwvJ2EN8IG_kgjsX6HQZ6dDhw/viewform

The use of AI-based tools is allowed.

По подсчётам во рту осталось 26 зубов, но половина всё равно запломбированная...