I'm delighted to release the first half of my new open-access online textbook in human population genetics: web.stanford.edu/group/pritchar…

Adaptive MCMC can be very useful in practice, but theoretical results are technical. We attempted to write a more accessible story about adaptive MCMC theory: arxiv.org/abs/2408.14903. It starts from the beautiful martingale decomposition of Andrieu & Moulines (2006):

It took me SEVEN years to figure out that you can save a ggplot as a vectorized .svg file ggsave("file.svg"). The svg file can be opened on powerpoint and it will be completely vectorized, even text are still recognized as text boxes! No more recreating plots on illustrator!!!!!

Announcing: AI in Industry and Academia Conversations ailab.criteo.com/ai-in-industry… Conversations with interesting people about machine learning and AI in industry and academia. Starting 19 September. Featuring: @DrMaggiori, Nicolas Chopin, Dawen Liang , flavianv , David Rohde,..

Last week to apply for machine learning postdoc positions in my team in Manchester or Helsinki

Newborn (extended abstract) paper in Bayesian Experimental Design (BED). This specific iteration takes the same approach as in arxiv.org/abs/2402.07868: formulate sequential planning as an amortized inference procedure, but differs on its computational treatment. 1/7

I made a new website: richardgeveritt.github.io It is mainly a Rosetta Stone type site, listing different online profiles (plus some software projects). Hopefully will be useful as social media continues to fragment...

ERC Starting Grant for Tom Berrett Tom Berrett has been announced as a recipient of an ERC Starting Grant worth €1.5 million. Information on Tom's research can be found on his website here: thomasberrett.github.io

This is absolute art. The field, the fading light, the celebrations…

Two new chapters from my free online book in human genetics out this weekend! These complete Part 3 of the book, on human population structure and history: 3.3: Inferring human prehistory from genetic data [this thread] 3.4: Ancient DNA [next thread] web.stanford.edu/group/pritchar…

#NeurIPS decisions are out! Congratulations to Jen Ning Lim (Warwick Statistics PhD Student) who has had a paper on Semi-implicit Variational Inference (preprint: arxiv.org/abs/2407.00649) accepted for a spotlight presentation at NeurIPS 2024.

Congratulations to David Huk, Yuanhe Zhang, Mark Steel, Ritabrata Dutta (all Warwick Statistics), whose paper "Quasi-Bayes meets Vines" (arxiv.org/abs/2406.12764) has been accepted for #NeurIPS ! In the paper they model high-dimensional densities with the Quasi-Bayesian Vine.

Do you think you know Maximum Likelihood? Think again It seems like such a natural idea, but there’s an epic and turbulent history with numerous assaults on the core idea, culminating in a beautiful and complicated theory. A highly entertaining account: arxiv.org/abs/0804.2996

Three of my PhD students have their viva in the next 18 days. "Go well" to all of them. I'll try to post a few things about their work in the next few weeks.

The One World Approximate Bayesian Inference (ABI) Seminar is on X. x.com/approxbayes

This is the new account for the One World Approximate Bayesian Inference (ABI) Seminar. Please follow if you are interested in any flavour of approximate Bayesian inference, including simulation-based inference. Website: warwick.ac.uk/fac/sci/statis…

The first talk of Season 6 is next week! On 31st October, 11am GMT, Ullrich Köthe (University of Heidelberg) will talk about "Free-form flows for physics-informed generative modelling". The talk will be on MS teams. Sign up to the list to get the link: listserv.csv.warwick.ac.uk/mailman/listin…

📢 Post-Bayesian online seminar series coming!📢 To stay posted, sign up at tinyurl.com/postBayes We'll discuss cutting-edge methods for posteriors that no longer rely on Bayes Theorem. (e.g., PAC-Bayes, generalised Bayes, Martingale posteriors, ...) Pls circulate widely!

At the University of Birmingham today, speaking to CENTA PhD students about "Understanding Machine Learning and Artificial Intelligence". centa.ac.uk