

Generative AI and Diffusion Models: a Statistical Physics Analysis (3/3)
By Giulio Biroli


Exploring the High-dimensional Random Landscapes of Data Science (3/3)
By Gérard Ben Arous
Appears in collection : 2016 - T1 - WS2 - Fundamental inequalities and lower bounds theme
Define the Stam region as the subset of the positive orthant in R^{2N−1} that arises from considering entropy powers of subset sums of N independent random vectors taking values in some Euclidean space. It is shown that the fractionally superadditive set functions give an outer bound for the Stam region, but that the supermodular set functions do not. In addition, various structural properties of the Stam region are explored.