-
MTA Rényi Intézet, kutyás tererm (harmadik emelet)
-
-
-
-
-
-

Description

 

Barycenters (or mean squared error estimators) play a distinguished role 
in statistics and information theory. This concept is boring in 
Euclidean spaces in the sense that it coincides with the weighted 
average. However, non-Euclidean metrics and other distance-like 
functions (such as relative entropies) are often more natural than the 
flat metric from the viewpoint of applications.
First, we take the submanifold of centered Gaussian measures in the 
space of square integrable random variables in R^d endowed with the 
optimal transport (Wasserstein) distance to illustrate the challenges of 
computing the barycenter in non-flat metrics. We will also discuss the 
closely related Riemannian trace metric on positive operators, which is 
defined by the Hessian of the Boltzmann entropy, from this viewpoint.
Then we turn to quantum information theory and consider generalized 
quantum Hellinger divergences, that belong to the family of maximal 
quantum f-divergences and behave like squared distances in some sense to 
be clarified during the talk.
We derive a characterization of the barycenters for these divergences 
and compare our results to those of Bhatia et al. [Lett. Math. Phys. 
(2019), in press, arXiv:1901.01378v1]. We note that the characterization 
given by Bhatia et al. is not correct in general, albeit it is true for 
commuting operators.
Based on joint work with Jozsef Pitrik (arXiv:1903.10455)