
My book about introductory data science was published today. It is available online - tellingstorieswithdata.com - and you can buy a print copy: routledge.com/Telling-Storie…. I would appreciate your feedback and suggestions for improvements.

I spend too much money on books. And too much time writing them. 🇦🇺. tellingstorieswithdata.com

My book about introductory data science was published today. It is available online - tellingstorieswithdata.com - and you can buy a print copy: routledge.com/Telling-Storie…. I would appreciate your feedback and suggestions for improvements.

Looking forward to your thoughts and comments: arxiv.org/pdf/2611.05548

2. You need to consistently keep that system updated (because if a submitter is using a better model yours might give bad advice), and you need to consistently judge the judges.

What does this all mean? 1. If you want to have AI act as an auto-reviewer then you need to make sure that you have multiple frontier models from different labs and run them multiple times for each review.

3. Advice from non-frontier models can actually make frontier model submissions worse. 4. There is still considerable self-bias - the models over-rate their own output.

A few other results: 1. Models can differ considerably from run to run - if you're using LLMs to evaluate something you need to run it multiple times 2. Some models still have no sense of self and sometimes name themselves as from another model provider!

I look at whether agentic AI can complete a data science project and whether that project is improved by other AI models giving feedback, and I find basically that such loops help weaker models but not stronger models - frontier performance is the frontier.

Very happy to have a pre-print of "Iterated creation, grading, and revision of data science projects by language models" available on arXiv.

Had a great visit to @UofT stats dept - thanks to @RohanAlexander & others for the invitation. Such a fantastic group!

What's the point of the academy, if our ideas cease to have influence and impact?

All welcome at the conference in 10 days in Toronto. And also if you think “hey I’ve got thoughts on AI and quant social science” and may like to contribute please email me.
I'll be at a workshop organized by @RohanAlexander next week on how AI will change quantitative social science. This is my short paper, and this is the key argument. ryancbriggs.net/blog/as-ai-l…

I'll be at a workshop organized by @RohanAlexander next week on how AI will change quantitative social science. This is my short paper, and this is the key argument. ryancbriggs.net/blog/as-ai-l…

I know nothing about physics, but it was interesting how similar the methodological concerns were to those we have in social science: theoryandpractice.org/2024/1…

In-person at 700 University Ave, Level 9, Room 9195 or Zoom (get link here: rohanalexander.substack.com/…).

This Friday 25 October 2024, noon (EDT), I'm thrilled to host Jay Alammar, Cohere, on his new book “Hands-On Large Language Models: Language Understanding and Generation”. All welcome.

Fresca absolutely the best in Toronto on a quality/price basis.
It's time for Toronto to stop chasing the hype of New York-style pizza, and to embrace our own homegrown classics. An ode to Fresca Pizza & Pasta, still the best slice in city. thestar.com/life/food-and-dr…