em algorithm – Optimization Online

Bounded perturbation resilience of projected scaled gradient methods

Published: 2015/07/27

Constrained Nonlinear Optimization, Convex Optimization em algorithm, perturbation resilience, projected scaled gradient methods, superiorization

We investigate projected scaled gradient (PSG) methods for convex minimization problems. These methods perform a descent step along a diagonally scaled gradient direction followed by a feasibility regaining step via orthogonal projection onto the constraint set. This constitutes a generalized algorithmic structure that encompasses as special cases the gradient projection method, the projected Newton method, … Read more

An Analysis of the EM Algorithm andEntropy-Like Proximal Point Methods

Published: 2001/12/17

Paul Tseng

Nonlinear Optimization convergence analysis, em algorithm, kullback-leibler distance, phi-divergence, proximal point algorithm

The EM algorithm is a popular method for maximum likelihood estimation from incomplete data. This method may be viewed as a proximal point method for maximizing the log-likelhood function using an integral form of the Kullback-Leibler distance function. Motivated by this interpretation, we consider a proximal point method using an integral form of entropy-like distance … Read more