Dear Colleagues,

as part of the Visiting Professors program of the Master's Degree in Stochastics and Data Science at the University of Torino (full program available at https://www.master-sds.unito.it/go/visiting), we are pleased to announce the following course:

--------------------------------------

Samuel LIVINGSTONE (UCL)

MONTE CARLO METHODS FOR STATISTICAL INFERENCE

Monte Carlo methods and sampling algorithms have had a profound effect on scientific research for the past 80 years. In the course we will discuss these approaches in the context of statistical inference. We will cover some mathematical preliminaries, the Monte Carlo method, and basic approaches to sample from a probability distribution in low dimensions, before motivating and introducing Markov chain Monte Carlo and the Metropolis—Hastings algorithm for high-dimensional sampling and Monte Carlo (MCMC). Afterwards we will discuss more advanced topics such as studying bias and variance of MCMC ergodic averages via mixing times, and more advanced algorithms such as adaptive and gradient-based methods.

--------------------------------------

The course will be held exclusively in person, according to the following schedule:

- Tue 7/5, h.14-16 - Wed 8/5, h.16-18 - Thu 9/5, h.16-18 - Tue 14/5, h.14-16 - Wed 15/5, h.16-18 - Thu 16/5, h.16-18

Course location: Room 11, third floor, Corso Unione Sovietica 218/bis, 10134, Torino.

The participation is open to all interested.

Best regards

--- Matteo Ruggiero University of Torino and Collegio Carlo Alberto www.matteoruggiero.it