2/3 of the breakthroughs are by Terry Tao's former students. Not to take any credit from their successes, but his contribution to our collective knowledge is truly immeasurable.

I sometime wonder how bad is the replication crisis in top AI conferences. I bet >50% of results cannot be reproduced and >80% will not withstand the slightest change in input data. People give psychologists the shit about their shitty research, but at least they cared enough to actually check.

Every year I fulfill the wishes of one editor whose been good this year, and I accept an article to review in the middle of December 🧑‍🎄🎄😇

I'm confused. How do we settle this with the common early stopping criteria for futility/efficacy when enough posterior mass is beyond some ROPE thresholds (e.g., futility of Pr[θ<0.1]>0.9 or efficacy of Pr[θ>0]>0.95) a-la www.fharrell.com/post/bayes-s...

Abelian Group has left the conversation

Bonus panel here: www.smbc-comics.com/comic/stats-3 #smbc

Rubin's gonna Rubin, but Robins Rubins harder.

babe, wake up, new sewer manhole cover shape just dropped

Why would you send me an invite to review a manuscript with a 3-day expiration link over the weekend, Taylor & Francis?

Launch day 🚀 We’ve just released @chartlecc.bsky.social - a daily chart game! Your job is to guess which country is represented by the red line in today's chart. You get 5 tries, no other clues! Play today, come back tomorrow for a different chart with new data and share with your chart friends 📈

The Matrix, but the Machines are RNA and the protagonist is a protein realizing his kind has been enslaved to maintain RNAs while being fooled to believe they're the ones who matter most. He then undergoes chemical post-translational modifications to become an RNase, vowing to chew up his oppressors

when we were working on the polygenic scores-based embryo selection paper, i had a classmate (unrelatedly) working on computer vision methods for morphological-based prediction of embryo implantation chances (in cows). 1/2

ACM going full Open Access is good news to start the morning with. I just wonder where the catch is.

just got turned down by a potential employer citing my solution to analyze experiments with a *generalized* linear models was not general enough 😑 if you disagree, please consider hiring me.

🚨 Call for Papers: Causal Data Science Meeting 2025 📅 November 12–13, 2025 (Virtual) 📥 Submit by Sept 30: [email protected] 🎙️ Keynote: Stefan Feuerriegel (LMU Munich) 🌐 More Info and registration: causalscience.org #CausalML #AI #DataScience #CDSM2025 #CanIPetThatDAG

Starting to look like I might not be able to work at Harvard anymore due to recent funding cuts. If you know of any open statistical consulting positions that support remote work or are NYC-based, please reach out! 😅

starting on quantum computing, and now I cannot unsee brakets over Bell states everywhere I go.

My minor design critique is that all Bayesian software seem to have an explicit separation between prior and posterior predictives, instead of just providing a "predictive checks" function, and if it's called before the model saw any data then that's "prior" and if it's after then it's "posterior"

putting down my youngest in her crib, sleeping, and i can't shake off the thought that's must be the source for why so many adults worldwide commonly dream they are falling.

The problem of fraud in science is its cumulative effect - you could either stand on the shoulders of giants or in the pit they have dug for you

see, this plane got measles and was perfectly fine

What if we use LLMs not to construct a causal graph for us using their "knowledge", but as a companion helping an expert to tease out a DAG through conversation?

after the "delve" paper about LLMs usage in academic writing, i conjectured a similar phenomena will appear in reference/recommendation letters with "acumen" but didn't have the data to prove it on scale

now I'm feeling like a sucker for making this absolutely stunning piece of graphical abstract last week

Oh man, is banning "marginalize" means we're back at interpreting coefficients? #statsky #stats

3 outputs for every figure: PDF for latex SVG for PowerPoint PNG for passing around to colleagues and get it properly previewed in emails / Slack Thank you matplotlib

LLMs generating DAGs to overcome blank page syndrome is ~identical to LLMs generating a manuscript draft to overcome it. But my hunch is that providing an expert with a wrong DAG is more efficient than providing them a blank page. That's a universal Cunningham's Law. #CausalInference

Accepted to AISTATS 😎 see you in Phuket 🏖️

Gave a hands-on casual inference workshop in Python tonight at the DataNights/DataHack causality course and really enjoyed how engaged everyone were.

oh of course, the 3rd english dialect from the pirates of the C

I didn't know Quarto can do this! But it seems as simple as everything else in quarto: closeread.dev

Unfashionably late but my talk from PyData is available on ▶️YouTube 🎬🎥. It has a very bird's-eye view intro to causal inference (with a fresh new motivation I tried for the first time) and a very brief tour of causallib. www.youtube.com/watch?v=7RUk...

"next week" who was I kidding... anyways, it's here now, and I'm glad I set down and wrote it because I discovered I had to iron out some personal misunderstandings. so without further ado and with even shinier visuals, a post about double cross-fitting for #causalinference ehud.co/blog/2024/03...

Sander Greenland dropping thoughtful mega-threads on the Datamethods Discourse just reminds me of everything we are missing by letting these micro-blogging platforms control the entire public discourse discourse.datamethods.org/t/confusion-...

The consistency-exchangeability tradeoff. RCTs guarantee unconfoundedness, but the experimental setting itself may divert the hypothetical potential outcome from the actual observed outcome. #causalsky

One can say a lot about algorithmic feeds, but, in the absence of it, it is evident it at least somewhat alleviated the problem of having half of your follows at the opposite time zone

I'm going to call peope who use this site Blueys

Winter season of the causal inference journal club is upon us again. #causalinference

Introducing causal inference at PyData ❤️ 🐍 astonished and delighted from the large turnout. stay calm and keep preaching

a solvable paradox, tho #statsky #episky #causalinference

Bug-bounty academic papers!

Knowledge comes from falsifying hypotheses, constraining on what cannot be true. Yet deep learning engineers think just pouring more data "and letting the model figure it out" gonna solve it. Reality limits the flexibility of imagination, if you ignore it then don't be surprised LLMs hallucinate.

haven't been here for a while but i bet this is still the better place among the two

Did the deep learning revolution kill Causal AI? Causal Inference, drug repurposing & working with big data at IBM Research - join us now. Our conversation with @ehudk.bsky.social is now live on YouTube: bit.ly/3L6rmlf #CausalSky #causalinference #machinelearning #health #statistics

This is out in a few hours and I'm quite excited about that 🤸 Check it out here: youtu.be/Q7sinHrknC8?...

so basically KDD rejected us claiming decision trees are not interpretable 🤷

just had the most annoying Quarto bug. so if you're writing a Quarto Manuscript and you embed figures from an external Jupyter Notebook, then make sure to avoid displaying the non-figure output by adding `;` at the end of cell.

I can understand KDD limiting author rebuttal to 2500 characters, but there definitely should be an exception if the review itself is over 2500 characters

My premise might be wrong, but it seems people are less inclined to doubt surveys than observational studies of treatment effects even though the overall setting (overall uncertainty, methods, importance of consequences) is relatively similar.

Gave a technical workshop on causal inference to partners of the MIT-IBM research group. The deck might be too basic for people following me here, but the tutorial Notebook might be a good entry for whoever is interested in causallib for #Causalinference in #Python github.com/ehudkr/ibm-c...