Regularized expectation maximization (RegEM) algorithm is an extension of the expectation-maximization (EM) algorithm developed by Tapio Schneider. It has been used in several multiproxy reconstructions being the current favorite of Michael E. Mann and his co-authors. Its use has become controversial, and the consensus at CA is that "The Team" uses RegEM because they like the results, not because they understand the statistics. See, for example, Smerdon et al 2008 on RegEM and other nearby threads.
RegEM is described in Schneider, T., 2001: Analysis of incomplete climate data: Estimation of mean values and covariance matrices and imputation of missing values, Journal of Climate, 14, 853–871. Matlab Implementation is available from Schneider's homepage.