Compressed Maximum Likelihood

                          Yi Hao 1 Alon Orlitsky 1

             Abstract                 An estimator for f is a function f : Z → Q such that when
   Maximum likelihood (ML) is one of the most          Z  p, the loss d(f(Z), f (p)) is small. The ML plug-in
   fundamental and general statistical estimation        estimator of f first finds the ML distribution estimate
   techniques. Inspired by recent advances in esti-
                                            ...