Jon Mellon

To make it worse, it then failed to fix the issue 4 times in a row

Gemini3-pro has a weirdly condescending personality.

Even my childless colleagues are all sick!

Does anyone else feel like there’s a lot more respiratory diseases going around this winter? Is there any data backing this up or is everyone I know unusually unlucky?

Well aware this makes me the biggest 37 year old dad cliche

This was the last book we read. Also a big thumb up from me

My favorite non-fiction book of all time

I loved the power broker but committing to the full LBJ books might be a tough sell for the group

Also very happy to read older books. Good histories have staying power

This looks really interesting!

Popular or academic suggestions welcome!

I’ve been running a history book club for the last few months. What are everyone’s favorite history books?

Definitely a totally unbiased list

Non fiction

Making of the atomic bomb
Against the grain
The vital question
The prize: the epic quest for oil money and power
Electoral shocks: the volatile voter in a turbulent world
Destiny disrupted: a history of the world through islamic eyes

Fiction
Hyperion
Children of time

It’s a real job! But sadly it actually involves things making classifications of diseases and medicines

I think there’s a case for teaching Bayesian stats as the default and all the frequentist procedures as helpful estimators for a special case of uniform priors

One day I want to create a wrapper package for lm called uniform_prior_lm that does nothing but relabel CIs and p-values as their Bayesian equivalents.

Does it behave well in the context of a few very large confounders?

I actually don’t know. I need to get more up to speed with double ML and how it relates to these problems. My gut reaction is that it wouldn’t solve it, but I trust simulations more than my gut

A meme with two panels. In panel 1 the woman looks disgusted. In panel 2 intrigued. Text, panel 1: fake data. Text, panel 2: silicon sample

I might just take you up on that. Will chat to Ben!

I would like to get back to this but that paper is currently in stasis

When I was 17 I really enjoyed the Pig that Wants to Be Eaten. No idea how it holds up

Camus clearly never had to work out what was wrong with district level outcomes at the 2007 Burkina Faso legislative elections

If you have two measures of the same confounder, you can just include both of them in your regression model Sometimes, researchers worry about multicollinearity in situations where it’s actually a non-issue. Here’s one such scenario. Imagine a situation where you are interested in the effect of X on Y (X...

New blog post!

Let's say you have two measures meant to capture the same confounder. They're highly correlated. Can you still proceed with your regression analysis?

(I admit, the title is a bit of a spoiler)

www.the100.ci/2025/10/13/i...

Has anyone had success stories (or can point me to writeups) with fine tuning frontier LLMs?

Education? Completed it, mate.

Last week, I defended my PhD thesis and am now a doctor!

Thank you to @robfordmancs.bsky.social and @jonmellon.bsky.social for being great supervisors for 4 years

And thanks to @sarahobolt.bsky.social and Ed Fieldhouse for their interesting feedback and discussion

These are continuous not discrete variables and every commentary on this is assuming different cutpoints

I would say that a natural parsing of “is it acceptable to rob a store for food?” either implies that you cannot get food in another way or is answered with an easy no