We present original empirical Bernstein inequalities for U-statistics with bounded symmetric kernels q. They are expressed with respect to empirical estimates of either the variance of q or the conditional variance that appears in the Bernsteintype inequality for U-statistics derived by Arcones [2]. Our result subsumes other existing empirical Bernstein inequalities, as it reduces to them when ...

Many real-world classification applications fall into the class of positive and unlabeled (PU) learning problems. In many such applications, not only could the negative training examples be missing, the number of positive examples available for learning may also be fairly limited due to the impracticality of hand-labeling a large number of training examples. Current PU learning techniques have ...

A fast learning algorithm for training multilayer feedforward neural networks (FNN's) by using a fading memory extended Kalman filter (FMEKF) is presented first, along with a technique using a self-adjusting time-varying forgetting factor. Then a U-D factorization-based FMEKF is proposed to further improve the learning rate and accuracy of the FNN. In comparison with the backpropagation (BP) an...

This paper introduces a network for volumetric segmentation that learns from sparsely annotated volumetric images. We outline two attractive use cases of this method: (1) In a semi-automated setup, the user annotates some slices in the volume to be segmented. The network learns from these sparse annotations and provides a dense 3D segmentation. (2) In a fully-automated setup, we assume that a r...

In this study, a SCORM-compliant ubiquitous learning grid was constructed by using the grid services technologies combined with CC/PP. The purpose was to let anyone access any information at anyplace, anytime, by any device to learn. Several SCORM-compliant learning management systems collaborated by Globus Toolkit 3.2 grid engine and CC/PP were implemented to provide a content adaptive environ...