Lightnews — Scholar-powered news

Nathaniel Forde

@nathanielforde.bsky.social

1K followers 720 following 180 posts

https://nathanielf.github.io/ Statistics, Probability previously Logic and Philosophy

nathanielf.github.io

Posts Media Videos Starter Packs

Pinned

Nathaniel Forde @nathanielforde.bsky.social · Jan 1

This last 12 months or so was intense and fun both professionally and personally. Thanks owed @benvincent.bsky.social , @twiecki.bsky.social, @alex-andorra.bsky.social and all the pymc devs! Looking forward to 2025 @pymc-devs.bsky.social My #pymc, #opensource year-in-review:

2 1 16

Reposted by Nathaniel Forde

Markus Eichhorn @markuseichhorn.bsky.social · 2d

Books Upstairs is a fine store with a particularly excellent poetry section. Why not support them by ordering your books online instead: booksupstairs.ie

The Irish Times @irishtimes.com · 2d

‘Really shocking’: Independent Dublin bookstore damaged in arson attack

Books Upstairs manager hopes to partially open the D’Olier Street store later on Monday

www.irishtimes.com

4 54 93

Reposted by Nathaniel Forde

juanitorduz @juanitorduz.bsky.social · 5d

It was fun (painful 😅) to implement VAR(p) models from scratch juanitorduz.github.io/var_numpyro/

Bayesian Vector Autoregressive Models in NumPyro - Dr. Juan Camilo Orduz

juanitorduz.github.io

1 3 14

Nathaniel Forde @nathanielforde.bsky.social · 7d

Congratulations!

Nathaniel Forde @nathanielforde.bsky.social · 11d

@benvincent.bsky.social

Nathaniel Forde @nathanielforde.bsky.social · 14d

@pymc-labs.bsky.social

Reposted by Nathaniel Forde

Henry Farrell @himself.bsky.social · 16d

The Aaron Swartz production function www.programmablemutter.com/p/the-aaron-...

The Aaron Swartz production function

Incalculable diffusion as a philosophy of doing

www.programmablemutter.com

2 35 100

Reposted by Nathaniel Forde

Kendra Albert @kendraserra.bsky.social · 27d

Very glad to see that someone is doing the important work of aligning AI alignment. alignmentalignment.ai

Center for the Alignment of AI Alignment Centers

We align the aligners

alignmentalignment.ai

22 110 410

Reposted by Nathaniel Forde

MIT Press @mitpress.bsky.social · 22d

"Open-mindedness is not incompatible with having opinions."

Not Every Idea Deserves Equal Time

A classic critique shows how creationists’ calls for “equal time” in classrooms blurred the line between legitimate scientific debate and intellectual imposture.

thereader.mitpress.mit.edu

1 6 26

Reposted by Nathaniel Forde

Huw Price @huwprice.bsky.social · 24d

Boosting this live tweeting of the excellent Emily Adlai’s new book to #philsci #philsky #quantum – thanks @jeffreymepstein.bsky.social for this public service! :)

Jeffrey M Epstein @jeffreymepstein.bsky.social · 29d

I just got a copy of this new book from Emily Adlam (with whom I did my masters, so this is the first book I'm reading written by a former classmate!). I'm going to try the thing where I tweet about it as I go.

global.oup.com

4 15

Nathaniel Forde @nathanielforde.bsky.social · 25d

Yeah, @tomicapretto.bsky.social also had an example with Bambi here: discourse.pymc.io/t/weights-fo...

Weights for Model in Bambi

Thanks for the example! I see now what you want (i.e. to weigh the contribution of each observation to the likelihood). I can think of a complex solution that I think solves your problem but it’s no...

discourse.pymc.io

Nathaniel Forde @nathanielforde.bsky.social · 27d

Thanks! Great to hear. Glad you liked it

Nathaniel Forde @nathanielforde.bsky.social · Sep 8

Super cool. Am actually diabetic, would be keen to do it. But not very prevalent in Ireland... did you get a prescription or is it over the counter in Germany?

1 1

Nathaniel Forde @nathanielforde.bsky.social · Sep 7

Cool pic. Run with it.

Reposted by Nathaniel Forde

Chelsea Parlett @chelseaparlett.bsky.social · Aug 29

I’m especially proud of this article I wrote about Gaussian Processes for the Recast blog! 🥳

GPs are super interesting, but it’s not easy to wrap your head around them at first 🤔

This is a medium level (more intuition than math) introduction to GPs for time series.

getrecast.com/gaussian-pro...

A Gaussian process showing that the allowed time series are forced to be compatible with data

2 24 78

Nathaniel Forde @nathanielforde.bsky.social · Aug 27

```On Monday, Pirro told a judge in a separate federal criminal case in D.C. that “an Indictment has not been returned” after “a __third__ grand jury returned a no true bill.”

The file drawer problem but for guilty verdicts

Olúfẹ́mi O. Táíwò @olufemiotaiwo.bsky.social · Aug 27

we are going to win

www.cnbc.com/2025/08/27/f...

Feds fail to get indictment against DC sandwich thrower fired by DOJ: Report

Sean Charles Dunn was fired from his job at the Department of Justice after the sub-chucking incident, which was caught on video.

www.cnbc.com

Reposted by Nathaniel Forde

PyMC Labs @pymc-labs.bsky.social · Aug 25

Market share predictions fail when models ignore how markets really work.

At #PyDataBerlin, @nathanielforde.bsky.social shows how PyMC Marketing’s new Consumer Choice module uses Bayesian models for realistic substitution.

🗓️ Sept 1, 4:20 PM | 📍 Room B05-B06
#BayesianStats #PyMC

1 2

Nathaniel Forde @nathanielforde.bsky.social · Aug 17

was more political than that... even rhetorically he would likely weaken the call to arms, were he to extoll on the benefits of protein folding. He wants to make the argument about the future of work, not the pitch deck of the next start up...

Nathaniel Forde @nathanielforde.bsky.social · Aug 17

Nearly every second idiot on linkedin frames themselves as a cassandra on this issue. The topic almost demands dystopian/utopian speculation.

But fair enough, I can see how he comes across as strident and purely negative. I just never read that as a lack of curiosity... just thought the writing

1 3

Nathaniel Forde @nathanielforde.bsky.social · Aug 17

the argument unmoderated. There is value in that even if the corrective swing overshoots. As to intellectual honesty, doesn't he wear his position on his sleeve? You can disagree with him... but his position is pretty clear...No?

1 2

Nathaniel Forde @nathanielforde.bsky.social · Aug 17

Not doubting that there are moderates on Blusky and I agree he's definitely towing a line and he's not giving weight in his articles to the positives or the wow factor of the tech... but, I mean that's fine. There are enough voices doing that already. His voice is of note because it puts forth

1 2

Nathaniel Forde @nathanielforde.bsky.social · Aug 17

I thought the rise of the business idiot essay seemed on point as a description of management style...

Also, I don't think anyone really understands the future value of AI. There is definitely room for a sceptical view of the upper estimates. Don't really see that to be controversial either...

1 1

Reposted by Nathaniel Forde

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · Aug 15

#rstats #statistics

I've released my new open source book, "Powered by Linear Algebra: the role of matrices and vector space in data science," at matloff.github.io/WackyLinearA....

Turns the classic LA course on its head! Still proves the theorems, but with a deep emphasis on applications.

matloff.github.io

10 24 84

Nathaniel Forde @nathanielforde.bsky.social · Aug 11

I'd be interested in reading an essay which traces a direct line from Krug's "Don't make me think" to Trump.

en.wikipedia.org/wiki/Don%27t...

Don't Make Me Think - Wikipedia

en.wikipedia.org

Nathaniel Forde @nathanielforde.bsky.social · Aug 10

😅 Bayesian multivariate nailed it fast, Simple VAE needed a ton of data, Complex VAE for missing data, just broke. Technically the SEM worked great (a little slow maybe)... but i kind of bundled it here in "Bayesian multivariate"
.

1 2

Nathaniel Forde @nathanielforde.bsky.social · Aug 10

The results?
✅ One nailed it fast
📈 One needed a ton of data
💥 One just broke
(Guess which was which?)

1 2