👻Spooky Jorge Loría 👻

It's that special, scary time of year -- the time of year when lots of academics grumble about having to complete innumerable training courses before the end of year

Counterfactuals are real, and they could’ve hurt you

This does not help the general dread of turning 30 this year

combinatorics are never on our side 😔

The abstract submission deadline for the ELLIS Workshop on Amortized Probabilistic ML has been extended to October 25th!

You still have time to submit a short abstract on your work in: Amortized Inference, Neural processes, Prior-fitted networks, Sequential Decision-Making, and Foundation models.

A slide with the text “Predicting a missing variate for a fully observed covariate, however, is not always the relevant predictive task.”, a probabilistic graphical model for predicting a missing variate given a partially observed covariate, and a corresponding equation for the posterior prediction distribution.

On Wed Dec 10 join me to learn about what a regression model is, and what a regression model is not, while raising funds for World Central Kitchen and United Farm Workers. For details on how to register, sponsored registration possibilities, and more see betanalpha.github.io/courses/.

ELLIS PhD Program: Call for Applications 2025 The ELLIS mission is to create a diverse European network that promotes research excellence and advances breakthroughs in AI, as well as a pan-European PhD program to educate the next generation of AI...

Interested in Bayesian workflow and cross-validation?
Join Aki Vehtari’s group at Aalto University as a PhD student!
Apply through the ELLIS PhD program by Oct 31 → ellis.eu/news/ellis-p...

Recent work: users.aalto.fi/~ave/publica...

"Where are your error bars?"

AI isn't replacing radiologists Radiology combines digital images, clear benchmarks, and repeatable tasks. But demand for human radiologists is at an all-time high.

In 2016 Hinton predicted that AI would replace all radiologists in five years. Ten years later, why hasn't it happened? This post is a great explainer.

www.understandingai.org/p/ai-isnt-re...

Chartle - A daily chart game Guess the country in red by analysing today's chart

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 📈

- Did I just bump my mug full of coffee on my desk?
Yes
- Did the mug do a really cool motion without spilling a drop?
Of course
- Do I feel my heart bumping like crazy?
Naturally

I did see that 84.7% of all statistics in social media are made up on the spot

BBC News headline: Frankenstein is monster success at Venice film festival

Actually, Frankenstein is a doctor success at the Venice film festival

- Future PI's, postdocs, PhD students: calls will follow
- Companies, researchers in AI and other fields: we are seeking collaboration partners
- Finland and Europe: multiplicative effects coming from ELLIS

Stay tuned for further calls and research results!

@ellisinstitute.fi @ellis.eu

Reminds me of a paper I saw in the last year that the limit of a decision tree can be made into a GP

I'm organizing the CUNY Math Ed/Comm Seminar for its second year ever!

Helsinki goes a full year without a traffic death A city traffic engineer credits the success to lower speed limits and smarter design.

HUGE NEWS FROM HELSINKI, FINLAND

Helsinki, a city of 690 000 people, has now had a full year without any traffic fatalities. None. No pedestrian, cyclist, scooter user or car deaths.

This is a huge achievement for a city of our size. 💪🏻

A little background 🧵 1/

yle.fi/a/74-20174831

it's only causal if it comes from the do calculus or potential outcomes regions, otherwise it's just sparkling out of sample prediction

Updating software be like:

Closed source: oh god please don't enshittify
Open source: oh god please don't break

Research Associate or Research Fellow In Machine Learning (2 Posts Available):Oxford Road

Postdocs and Research Fellows: Probabilistic machine learning, Centre for AI Fundamentals, Manchester, UK, DL July 27, 2025

www.jobs.manchester.ac.uk/Job/JobDetai...

Funded by UKRI Turing AI World-Leading Researcher Fellowship, in the Manchester Centre for AI Fundamentals.

According to the pigeonhole principle, we really need to invest in pigeon housing.

Departure is now delayed 1 h 20 mins, because of them. I should've booed with the rest

Someone arrived late to the flight, which delayed the departure, and they got booed

A short blogpost detailing my experience of censorship at the AI for Good Summit with links to both original and censored versions of slides and links to my talk

aial.ie/blog/2025-ai...

*sighs* 1.96

Innovation thrives in inclusive environments! The study from @aalto.fi and @univaasa.bsky.social spanned 2003–2017, and more recent data indicates that the trend is accelerating. Seems like a good time to remind the world that inclusivity gives ideas room to flourish. 🏳️‍🌈
www.aalto.fi/en/news/lgbt...

It's that time of year again! Drowning doesn't look like drowning; familiarize yourself with the instinctive drowning response

If I were queen of the internet, you’d be able to write LaTeX anywhere, in any editor/textbox/message and it would render in line perfectly

Reached that point of coding where I search for an error and the answers I find online are "I asked [some LLM] and it helped" followed with a completely useless block of code 😔

I'm training an SVM to classify correct/incorrect proofs of suitably encoded mathematical proofs, such as Fermat's Last Theorem's.

Unsuccessfully so far, the margin is too small