RESUMEN
The 2015 European Guidelines on Diagnosis and Treatment of Pulmonary Hypertension are also valid for Germany. The guidelines contain detailed recommendations for the targeted and supportive treatment of pulmonary arterial hypertension (PAH). However, the practical implementation of the European Guidelines in Germany requires the consideration of several country-specific issues and already existing novel data. This requires a detailed commentary to the guidelines, and in some aspects an update already appears necessary. In June 2016, a Consensus Conference organized by the PH working groups of the German Society of Cardiology (DGK), the German Society of Respiratory Medicine (DGP) and the German Society of Pediatric Cardiology (DGPK) was held in Cologne, Germany. This conference aimed to solve practical and controversial issues surrounding the implementation of the European Guidelines in Germany. To this end, a number of working groups was initiated, one of which was specifically dedicated to general and supportive therapy of PAH. This article summarizes the results and recommendations of this working group.
Asunto(s)
Cardiología/normas , Hipertensión Pulmonar/diagnóstico , Hipertensión Pulmonar/terapia , Guías de Práctica Clínica como Asunto , Neumología/normas , Antihipertensivos/uso terapéutico , Determinación de la Presión Sanguínea/normas , Terapia Combinada/normas , Endarterectomía/normas , Alemania , HumanosRESUMEN
Proper initialization is one of the most important prerequisites for fast convergence of feedforward neural networks like high-order and multilayer perceptrons. This publication aims at determining the optimal variance (or range) for the initial weights and biases, which is the principal parameter of random initialization methods for both types of neural networks. An overview of random weight initialization methods for multilayer perceptrons is presented. These methods are extensively tested using eight real-world benchmark data sets and a broad range of initial weight variances by means of more than 30000 simulations, in the aim to find the best weight initialization method for multilayer perceptrons. For high-order networks, a large number of experiments (more than 200000 simulations) was performed, using three weight distributions, three activation functions, several network orders, and the same eight data sets. The results of these experiments are compared to weight initialization techniques for multilayer perceptrons, which leads to the proposal of a suitable initialization method for high-order perceptrons. The conclusions on the initialization methods for both types of networks are justified by sufficiently small confidence intervals of the mean convergence times.
RESUMEN
The backpropagation algorithm is widely used for training multilayer neural networks. In this publication the gain of its activation function(s) is investigated. In specific, it is proven that changing the gain of the activation function is equivalent to changing the learning rate and the weights. This simplifies the backpropagation learning rule by eliminating one of its parameters. The theorem can be extended to hold for some well-known variations on the backpropagation algorithm, such as using a momentum term, flat spot elimination, or adaptive gain. Furthermore, it is successfully applied to compensate for the nonstandard gain of optical sigmoids for optical neural networks.