The stability of the mold level in continuous slab casting machines is an important factor determining the final quality of flat carbon steel.The existing performant feedback control system of Slab Caster 2 in the Steel Shop of ArcelorMittal Gent will be extended with a feedforward signal using the predicted mold level.A prediction algorithm, an early warning detection system and detailed FFT analysis of the mold level variations have been designed in MATLAB-Simulink. PLC code is generated directly from the MATLAB design, and will be implemented in a PLC serving as PROFINET I-Device in the base automation of slab caster KG2.

In current healthcare environments, a trend toward mobile and personalized interactions between people and nurse call systems is strongly noticeable. Therefore, it should be possible to locate patients at all times and in all places throughout the care facility. This paper aims at describing a method by which a mobile node can locate itself indoors, based on signal strength measurements and a minimal amount of yes/no decisions. The algorithm has been developed specifically for use in a healthcare environment. With extensive testing and statistical support, we prove that our algorithm can be used in a healthcare setting with an envisioned level of localization accuracy up to room revel (or region level in a corridor), while avoiding heavy investments since the hardware of an existing nurse call network can be reused. The approach opted for leads to very high scalability, since thousands of mobile nodes can locate themselves. Network timing issues and localization update delays are avoided, which ensures that a patient can receive the needed care in a time and resources efficient way.

It has been shown that kernel based regression (KBR) with a least squares loss has some undesirable properties from robustness point of view. KBR with more robust loss functions, e.g. Huber or Logistic losses, often give rise to more complicated computations. In classical statistics, robustness is improved by reweighting the original estimate. We study the influence of reweighting the LS-KBR estimate using three well-known weight functions and one new weight function called Myriad. Our results give practical guidelines in order to choose the weights, providing robustness and fast convergence. It turns out that Logistic and Myriad weights are suitable reweighting schemes when outliers are present in the data. In fact, the Myriad shows better performance over the others in the presence of extreme outliers (e.g. Cauchy distributed errors). These findings are then illustrated on toy example as well as on a real life data sets. Finally, we establish an empirical maxbias curve to demonstrate the ability of the proposed methodology.

Bias-corrected approximate 100(1-α)% pointwise and simultaneous confidence and prediction intervals for least squares support vector machines are proposed. A simple way of determining the bias without estimating higher order derivatives is formulated. A variance estimator is developed that works well in the homoscedastic and heteroscedastic case. In order to produce simultaneous confidence intervals, a simple Šidák correction and a more involved correction (based on upcrossing theory) are used. The obtained confidence intervals are compared to a state-of-the-art bootstrap-based method. Simulations show that the proposed method obtains similar intervals compared to the bootstrap at a lower computational cost.

This paper studies a distribution-free estimator of the conditional support and tolerance intervals of a distributions underlying a set of paired independent and identically distributed (i.i.d.) observations. The key ingredients are (a) an appropriate notion of risk which measures what probability mass is not captured by the estimate, (b) a uniform concentration inequality for the empirical risk based on a compression argument, and (c) the derivation of a lower bound to the mutual information, dictating how to maximize the informativeness of the estimator. For this result we extend Fano's inequality to the bivariate case.

This paper proposes primal-dual kernel machine classifiers based on worst-case analysis of a finite set of observations including missing values of the inputs. Key ingredients are the use of a componentwise support vector machine (cSVM) and an empirical measure of maximal variation of the components to bind the influence of the component which cannot be evaluated due to missing values. A regularization term based on the L/sub 1/ norm of the maximal variation is used to obtain a mechanism for structure detection in that context. An efficient implementation using the hierarchical kernel machines framework is elaborated.