The war will be over by Christmas.
Maybe even this year.
Posts by Thomas Lumley
Our team went through the data file and removed the most obviously false responses — the impossible or extremely improbable ages, the ethnic/racial mismatches, the people who gave AI-generated free-response answers. When it was all said and done, we removed 574 of the 1,174 current Orthodox (48.9%) and 422 of the 695 former Orthodox (70.8%). Combined, we removed 53.3% of the respondents we were sent — and remember, that’s after a bunch of bad respondents were already removed in the data cleaning process prior to our receipt of the data.
Unfortunately, a great number of bogus respondents can sneak into opt-in online surveys, polluting the results. Among Orthodox Christians in a recent opt-in survey, about half seemed to be bogus respondents.
orthodoxstudies.substack.com/p/what-the-quiet-revival...
In a new preprint, the authors identify 124 peer-reviewed papers which used provenance-unknown Kaggle data to train AI for stroke and diabetes risk. Some sets show impossible patterns, yet models of the models are already hitting clinics.
#MedSky #MedAI
The Mastodon client I use lets you keep private notes about accounts, but I don't know if those notes are retained for blocked accounts
Not even that-- more just civic hygiene
It's probably not the explanation in your case, but I do block quite a lot of people I have never interacted with, mostly for being dicks in someone else's replies.
I'm sure they're wrong about *something*
All models are wrong; many models are useless
"A poem should not mean, but be"
Rare sighting of a column chart taking a dip at Hove Beach.
#dataviz
Springer maths books are famous for this
A particular sort of chemist. I prefer Derek Lowe's "The straight liquid must really be a treat, but I've never seen it in that form, and would only wish to through binoculars'
"Unlike hydrogen fluoride, anhydrous liquid hydrogen bromide is difficult to work with as a solvent" -wikipedia
Unlike?!!
"agentify" has been consultificated
Agree completely. This looks like a variation on good ol' staurosporine, which hits half the kinases known to man. This compound is great news, but surely not because it's some sort of selective inhibitor.
I'm pretty sure it would be valid even if it might not be licit
I'm not going to be impressed until they make machines that can fly
🚴♂️
A kauri tree looking towards the canopy.
PhD opportunity- fully funded with stipend, fees and project costs covered. Explore the role of phosphite in treating kauri dieback disease. Applications due 8th May. Please spread the word!
newzealandecology.org/phd-explorin...
That's not a consistent estimator of the true covariance matrix of Y, just an unbiased one
#nerd #nerder #nerdest
The new nominee for head of CDC is "a well-qualified former public health official and board-certified physician in preventive medicine, who has publicly supported vaccination and followed evidence-based medicine." And also a Black woman.
#WeirdlyNormal
Unless you're custard
There was never good evidence for it
the most interesting thing about this story is not the survey itself—which is a year old and has since been retracted over huge problems with the sample—but that media outlets keep trying to push stories of a massive religious revival among the young in absence of any evidence
Less bad? But, I mean, there's no question the US could destroy Iran if they really wanted, and killing the president seems like it might make them really want to.
Yes, but attacking the president on US soil doesn't seem likely to be strategically helpful.
It's the transition from pluvial to fluvial that you need to worry about
A newspaper cartoon with two grey-type aliens talking to a man startled by them while raking leaves in a garden. A flying saucer hovers behind the aliens, beaming light downwards. The aliens are saying "Nobody pick up their phone anymore, so we're back to face-to-face data collection."