Two nonlinear Kalman smoothers are proposed using the Student’s t distribution. The first, which we call the T-Robust smoother, finds the maximum a posteriori (MAP) solution for Gaussian process noise and Student’s t observation noise. It is extremely robust against outliers, outperforming the recently proposed `1-Laplace smoother in extreme situations with data containing 20% or more outliers....

In this paper, finite impulse response (FIR) smoothers are proposed for discrete-time systems. The proposed FIR smoother is designed under the constraints of linearity, unbiasedness, FIR structure, and independence of the initial state information. It is also obtained by directly minimizing the performance criterion with unbiased constraints. The approach to the MVF smoother proposed in this pa...

In the present paper we concentrate on an important issue in constructing a good multigrid solver: the choice of an efficient smoother. We will introduce all-at-once multigrid solvers for optimal control problems which show robust convergence in the grid size and in the regularization parameter. We will refer to recent publications that guarantee such a convergence behavior. These publications ...

Anat Sakov, Ilan Golani, Dina Lipkind, Yoav Benjamini Tel Aviv University Abstract In recent years, a growing need has arisen in different fields, for the development of computational systems for automated analysis of large amounts of data (high-throughput). Dealing with non-standard noise structure and outliers, that could have been detected and corrected in manual analysis, must now be built ...

The Ordinary Least Squares (OLS) method is the most popular technique in statistics and is often use to estimate the parameters of a model because of tradition and ease of computation. The OLS provides an efficient and unbiased estimates of the parameters when the underlying assumptions, especially the assumption of contant error variances (homoscedasticity), are satisfied. Nonetheless, in real...

The median absolute deviation (MAD) of each trait (principal component) was used to measure within-line variance, as it is less sensitive to non-normality than other measures of spread [67, 68]. However, plotting the raw MADs of each line against the line median tended to show a dependency of the MAD on the median (Figure S2). To correct for this dependency, the residuals of a local (lowess) re...