Botond Szabo (Bocconi University)

On distributed estimation and testing

In recent years, the amount of available information has become so vast in certain fields of applications that it is infeasible or undesirable to carry out all the computations on a single server. This has motivated the design and study of distributed statistical or learning approaches. In distributed methods, the data is split amongst different administrative units and computations are carried out locally in parallel to each other. The outcome of the local computations are then aggregated into a final result on a central machine.

In this talk we will consider the limitations and guarantees of distributed methods under communication constraints (i.e. only limited amount of bits are allowed to be transmitted between the machines) in context of the many normal means and Gaussian white noise model. We derive minimax lower bounds, matching upper bounds both for testing and estimation.

(joint work with Harry van Zanten and Lasse Vuursteen)

This talk will be held fully online; you are welcome to attend the talk via Zoom via the link at the bottom of the invitation text.