Learn More
Deep neural networks have recently become the gold standard for acoustic modeling in speech recognition systems. The key computational unit of a deep network is a linear projection followed by a point-wise non-linearity, which is typically a logistic function. In this work, we show that we can improve generalization and make training of deep networks faster(More)
— The QoS Steiner Tree Problem asks for the most cost-efficient way to multicast multimedia to a heterogeneous collection of users with different consumption rates. We assume that the cost of using a link is not constant but rather depends on the maximum bandwidth routed through the link. Formally, given a graph with costs on the edges, a source node and a(More)
In this paper, an anatomically accurate three-dimensional finite-element (FE) model of the human lumbar spine (L2-L3) was used to study the biomechanical effects of graded bilateral and unilateral facetectomies of L3 under anterior shear. The intact L2-L3 FE model was validated under compression, tension, and shear loading and the predicted responses(More)
The efficiency of real-time digital image processing operations has an important impact on the cost and realizability of complex algorithms. Global motion estimation is an example of such a complex algorithm. Most digital image processing is carried out with a precision of 8 bits per pixel, however there has always been interest in low-complexity(More)
Pressure-induced amorphous-to-amorphous configuration changes in Ca-Al metallic glasses (MGs) were studied by performing in-situ room-temperature high-pressure x-ray diffraction up to about 40 GPa. Changes in compressibility at about 18 GPa, 15.5 GPa and 7.5 GPa during compression are detected in Ca(80)Al(20), Ca(72.7)Al(27.3), and Ca(66.4)Al(33.6) MGs,(More)