Search results for "Function"

Neural Networks with Multidimensional Cross-Entropy Loss Functions

2019

Deep neural networks have emerged as an effective machine learning tool successfully applied for many tasks, such as misinformation detection, natural language processing, image recognition, machine translation, etc. Neural networks are often applied to binary or multi-class classification problems. In these settings, cross-entropy is used as a loss function for neural network training. In this short note, we propose an extension of the concept of cross-entropy, referred to as multidimensional cross-entropy, and its application as a loss function for classification using neural networks. The presented computational experiments on a benchmark dataset suggest that the proposed approaches may …

Global exponential stability of delayed Markovian jump fuzzy cellular neural networks with generally incomplete transition probability

2014

The problem of global exponential stability in mean square of delayed Markovian jump fuzzy cellular neural networks (DMJFCNNs) with generally uncertain transition rates (GUTRs) is investigated in this paper. In this GUTR neural network model, each transition rate can be completely unknown or only its estimate value is known. This new uncertain model is more general than the existing ones. By constructing suitable Lyapunov functionals, several sufficient conditions on the exponential stability in mean square of its equilibrium solution are derived in terms of linear matrix inequalities (LMIs). Finally, a numerical example is presented to illustrate the effectiveness and efficiency of our res…

Efficient MLP Digital Implementation on FPGA

2005

The efficiency and the accuracy of a digital feed-forward neural networks must be optimized to obtain both high classification rate and minimum area on chip. In this paper an efficient MLP digital implementation. The key features of the hardware implementation are the virtual neuron based architecture and the use of the sinusoidal activation function for the hidden layer. The effectiveness of the proposed solutions has been evaluated developing different FPGA based neural prototypes for the High Energy Physics domain and the automatic Road Sign Recognition domain. The use of the sinusoidal activation function decreases hardware resource employment of about 32% when compared with the standar…

Semi-Supervised Support Vector Biophysical Parameter Estimation

2008

Two kernel-based methods for semi-supervised regression are presented. The methods rely on building a graph or hypergraph Laplacian with both the labeled and unlabeled data, which is further used to deform the training kernel matrix. The deformed kernel is then used for support vector regression (SVR). The semi-supervised SVR methods are sucessfully tested in LAI estimation and ocean chlorophyll concentration prediction from remotely sensed images.

Regularized RBF Networks for Hyperspectral Data Classification

2004

In this paper, we analyze several regularized types of Radial Basis Function (RBF) Networks for crop classification using hyperspectral images. We compare the regularized RBF neural network with Support Vector Machines (SVM) using the RBF kernel, and AdaBoost Regularized (ABR) algorithm using RBF bases, in terms of accuracy and robustness. Several scenarios of increasing input space dimensionality are tested for six images containing six crop classes. Also, regularization, sparseness, and knowledge extraction are paid attention.

Classification of Satellite Images with Regularized AdaBoosting of RBF Neural Networks

2008

The AVL-mode: a safe closed loop algorithm for ventilation during total intravenous anesthesia.

1994

The Adaptive Lung Ventilation Controller (ALV-Controller) represents a new approach to closed loop control of ventilation. It is based on a pressure controlled ventilation mode. Adaptive lung ventilation signifies automatic breath by breath adaptation of breathing patterns to the lung mechanics of an individual patient. The specific goals are to minimize work of breathing, to maintain a preset alveolar ventilation and to prevent the occurrence of intrinsic PEEP. We ventilated 5 patients undergoing major abdominal procedures using ALV. ALV was tolerated well in all patients. Alveolar ventilation was preset between 5500 and 6500 ml/min. Serial dead space (Vds) and respiratory time constant (r…

Successful treatment of a patient with ARDS after pneumonectomy using high-frequency oscillatory ventilation.

1999

High frequency oscillatory ventilation (HFOV) was used in a patient who developed the acute respiratory distress syndrome 5 days following a right pneumonectomy for bronchogenic carcinoma. When conventional pressure-controlled ventilation failed to maintain adequate oxygenation, HFOV dramatically improved oxygenation within the first few hours of therapy. Pulmonary function and gas exchange recovered during a 10-day period of HFOV. No negative side effects were observed. Early use of HFOV may be a beneficial ventilation strategy for adults with acute pulmonary failure, even in the postoperative period after lung resection.

Enzyme replacement therapy for mucopolysaccharidosis VI: evaluation of long-term pulmonary function in patients treated with recombinant human N-acet…

2010

Pulmonary function is impaired in untreated mucopolysaccharidosis type VI (MPS VI). Pulmonary function was studied in patients during long-term enzyme replacement therapy (ERT) with recombinant human arylsulfatase B (rhASB; rhN-acetylgalactosamine 4-sulfatase). Pulmonary function tests prior to and for up to 240 weeks of weekly infusions of rhASB at 1 mg/kg were completed in 56 patients during Phase 1/2, Phase 2, Phase 3 and Phase 3 Extension trials of rhASB and the Survey Study. Forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1) and, in a subset of patients, maximum voluntary ventilation (MVV), were analyzed as absolute volume in liters. FEV1 and FVC showed little change f…

MR2553995 (2010h:26008): Mihail, Alexandru The Arzela-Ascoli theorem for partial defined functions. An. Univ. Bucureşti Mat. 57 (2008), no. 2, 259–26…

2008

In this paper the author gives a generalization of the Arzela-Ascoli theorem for partial defined functions, i.e., for functions defined in a nonempty subset of a metric space X and taking values in a metric space Y. To this end suitable definitions of local and uniform convergence for partial defined functions are introduced. As an application a different proof of a known result concerning the existence of Lipschitz selections for Lipschitz multifunctions is given. Reviewed by Luisa Di Piazza