Daniela Cialfi's contribution to the Discussion of 'the Discussion Meeting on Probabilistic and statistical aspects of machine learning'.

The article titled "From Denoising Diffusions to Denoising Markov Models" explores the use of denoising diffusion models in two ways: transforming data distribution into a Gaussian one and simulating approximate posterior simulation. The authors propose a framework that extends this approach to a wide range of spaces and introduces an original extension of score matching. The article suggests testing these models in real scenarios, such as theoretical neurobiology, to further explore their mathematical aspects. The author of the article is Daniela Cialfi. [Extracted from the article]