Joseph Lam (Magdeburg)
The consequences of local differential privacy constraints on goodness-of-fit testing are considered, i.e. the statistical problem assessing whether sample points are generated from a fixed density or not. The observations are hidden and replaced by a stochastic transformation satisfying the local differential privacy constraint. The focus will be on the lower bound, leading to the minimax optimality of our result over Besov balls.
The Zoom access data are available on the programm of the Research Seminar. See
www.math.uni-potsdam.de/professuren/wahrscheinlichkeitstheorie/forschung/forschungsseminar
