Home Statistics and Visualization Statistical Modeling, Causal Inference, and Social Science Online seminar for Monte Carlo Methods++

Online seminar for Monte Carlo Methods++

Statistical Modeling, Causal Inference, and Social Science 2024-09-27

This is Bob.

There’s a new weekly online seminar on all aspects of Monte Carlo (including adjacent topics like generative modeling and uncertainty quantification). Here’s the home page for the seminar.

The website includes the schedule, and links for Zoom, the Google mailing list, and the YouTube channel. Their site states that seminars take place on

  • Tuesdays at 8:30 am PT / 11:30 am ET / 4:30 pm London / 5:30 pm Paris / 11:30 pm Beijing

Presumably they’ll announce new times to cope with asynchronous daylight time changeovers.

They’re starting with an A-list of speakers:

  • Persi Diaconis – 10/01/2024
  • Mike Giles – 10/08/2024
  • Art Owen – 10/15/2024 (Art just gave a practice version of this introduction to quasi Monte Carlo here at Flatiron and it was really great)
  • Gareth Roberts – 10/29/2024 (I recently saw Gareth talk about his solution to the federated learning problem with MCMC using diffusions and Brownian bridges—Art’s talk included a really nice intro to Brownian bridges at a level even I could understand)

Mike Giles wrote the fundamental matrix autodiff results we used for Stan—I didn’t know he worked on sampling. Art Owen gave a practice version of his talk at Flatiron Institute yesterday and it included crystal clear introductions to several ideas I’d never seen before including multiple ways to generate quasi Monte Carlo points, Brownian bridges (which I’d never seen introduced in a way I could understand), etc.

The seminars are being organized by Valentin De Bortoli (DeepMind), Yang Chen (Michigan), Nianqiao Ju (Purdue), Sifan Liu (Flatiron/Duke), Sam Power (Bristol), Qian Qin (Minnesota), and Guanyang Wang (Rutgers).