Paul Hünermund

Counterintuitive but makes sense: wfh doesn’t encourage a “work from anywhere” world, rather it reinforces large cities bc long commute times become less of a disutility

A reminder that the 70+ year-old research discipline of AI is much bigger than genAI, much bigger than just deep learning, much bigger than Big Tech. AI includes symbolic reasoning, multi-agent systems and statistical machine learning and is used for astrophysics, renewable energy and climate models

I make a cross-over once in a while 😉

Meta information in the supplementary Word file 🙄

And yes, it immediately came back. But luckily only because of some formatting issues. 😅

YouTube video by UltimateTiktoks Basketball Tiktok Fails

Submitted a paper today. Every time I have this TikTok in my head. youtube.com/shorts/6kllg...

Twitter seems like an endless monologue from Tom Buchanan about the "rise of the Colored empires"

Stephen Nunez @socio-steve.bsky.social • 13h The AK in AK-47 stands for "Assault Knife" Steve @rapidlydownhill.bsky.social • 3h No. The AK-47, or Avtomat Kalashnikova model of the year 1947. Designed by Mikhail Kalashnikov. Avtomat translates to Automatic (or contextually to Automatic Machine). Kalashnikova means it "belongs" or is "of" Kalashnikov, the designer. Stephen Nuñez @socio-steve.bsky.social • 2h No it's assault knife. Just like how the AR in AR-15 stands for assault rifle. Steve @rapidlydownhill.bsky.social • 2h Still unconvinced? Stephen Nuñez @socio-steve.bsky.social • 2h Look is assault rifle and assault knife. Does exactly what is says on the tin. You Internet weirdos make everything more complicated than it needs to be. Steve @rapidlydownhill.bsky.social • 2h So. No acceptance of facts? Fair enough. You can believe what you want. I might have expected more from an apparently well educated person. Expect others to learn but not be willing to do so yourself. Resorting to insults is another sign of your insecurity. Good day sir.

Because historical counterfactuals are hard to pin down for most people, and the resulting epistemological uncertainty is frequently supplanted by prevailing culture war narratives.

At 17, Hannah Cairo Solved a Major Math Mystery | Quanta Magazine After finding the homeschooling life confining, the teen petitioned her way into a graduate class at Berkeley, where she ended up disproving a 40-year-old conjecture.

What a fantastic accomplishment -- and what a fantastic story! www.quantamagazine.org/at-17-hannah...

1/ An 'army mafia' has developed within Russia's invasion force in Ukraine, operating with near-impunity to smuggle commodities back into Russia and strip seized industries to sell for personal profit. A Russian commentary highlights the difficulties of tackling it. ⬇️

I'm skeptical whether economists will be able to lead the revolution to causal in industry. They are too deeply invested in their quasi-experimental methods and by now seem to be too narrow-minded to change.

🗣️ @aligarciaherrero.bsky.social joins Théo Storella & @yuyunzhan.bsky.social for a podcast discussion on the evolving landscape for European companies operating in China and their shifting strategies amid rising economic uncertainty and geopolitical tension. #EconSky
🔗 Listen here: buff.ly/Z3aJAWB

That's a good point. It's exactly the problem of eliciting expert domain knowledge from diverse stakeholders ("no causes in, no causes out") that we're currently working on. Causal AI is more demanding in that regard than simply running XGBoost or using PyTorch out of the box.

Double machine learning and automated confounder selection: A cautionary tale Double machine learning (DML) has become an increasingly popular tool for automated variable selection in high-dimensional settings. Even though the ability to deal with a large number of potential co...

Here: using DAGs to show how dangerous DML is www.degruyterbrill.com/document/doi...

It'll fade. DML is just too convenient to sanctify your regression models using the big data/ML moniker with.

Why not gonna happen?

Dann können wir uns wenigstens weiter überlegen gegenüber den unwissenden Normies fühlen.

Good, that gives me confidence. Bayesian meta-analysis won't be the next big thing, though.

Me preparing for the next AI winter—from which #CausalAI will emerge as the next big thing

Not following the discourse too much, but "Sydney Sweeney has good jeans" is a bad slogan, isn't it?

That's why a new Bond will have such a hard time to revitalize the franchise. Britain has lost too much of its cool.

to be clear, australian voters would not support this because one of our deep deficits as a nation is that we dont believe in the value of intellectual ambition in any way, this is not a case of albo being stupid (there are many of those but) we as a group are stupid

It's not nice to compare Judea with Yoda

Dude.

Big thanks to Professors Katz, Abraham, Dynan, and Troske for this wise and quick response.

Well played.

Would the Jedi or the Sith be more likely to develop Bayesian statistics?

In Europe

Bluesky doesn't seem to be the right place to get an unbiased answer to that.