removal of artifacts from bio-signals is a necessary step before automatic processing and obtaining clinical information. recently, many applications of empirical mode decomposition (emd) on biomedical researches have been presented that artifact reduction from bio-signals is one of them. emd separates a time series into finite numbers of its individual oscillations, which are called intrinsic ...

3 Convergence proof 4 3.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.2 Some basic inequalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.3 A bound on the suboptimality bound . . . . . . . . . . . . . . . . . . . . . . 7 3.4 A stopping criterion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.5 Numerical examp...

We present a method for the computation of stopping criterion for linear classical iterations in inexact aane-invariant Newton techniques. We show that the same methodology does not hold for non-classical iterative methods.

This paper focuses on linear classification using a fast and simple algorithm known as the Ho–Kashyap learning rule (HK). In order to avoid overfitting and instead of adding a regularization parameter in the criterion, early stopping is introduced as a regularization method for HK learning, which becomes HKES (Ho–Kashyap with Early Stopping). Furthermore, an automatic procedure, based on genera...

A soft-input soft-output decoder based on the least mean square error (LMSE) criterion is presented and employed for iterative decoding of product codes. The residual estimation error is derived and used for analyzing the convergence of the iterative process as well as a stopping criterion. A relation between the LMSE and the MAP decoders is also established. Simulation results are presented fo...

| By means of the mean reliability (de-ned as the mean of the absolute values of log-likelihood ratios), a new design of parallel con-catenated \turbo" codes is proposed. This criterion allows us to describe the behavior of the constituent decoders and, furthermore, to predict the behavior of the iterative decoder for large block lengths. The mean reliability can also be used as a stopping crit...

This thesis examines the use of support vector machines for active learning using linear, polynomial and radial basis function kernels. In our experiments we used named entity recognition which was treated as a binary task and as a multiclass task and we also tackled shallow parsing. We report savings in annotation costs ranging from 80% to 95% depending on the task. We observed that the distri...

We study the switch distribution, introduced by van Erven, Grünwald and De Rooij (2012), applied to model selection and subsequent estimation. While switching was known to be strongly consistent, here we show that it achieves minimax optimal parametric risk rates up to a log logn factor when comparing two nested exponential families, partially confirming a conjecture by Lauritzen (2012) and Cav...