This article demonstrates that we can apply deep learning to text understanding from characterlevel inputs all the way up to abstract text concepts, using temporal convolutional networks(LeCun et al., 1998) (ConvNets). We apply ConvNets to various large-scale datasets, including ontology classification, sentiment analysis, and text categorization. We show that temporal ConvNets can achieve asto...

A fascinating and deep question about nature is what one would see if one could probe space and time at smaller and smaller distances. Already the 19th-century founders of modern geometry contemplated the possibility that a piece of empty space that looks completely smooth and structureless to the naked eye might have an intricate microstructure at a much smaller scale. Our vastly increased und...

This thesis presents a deep neural network for word-level machine translation quality estimation. The model extends the feedforward multi-layer architecture by [Collobert et al., 2011] to learning continuous space representations for bilingual contexts from scratch. By means of stochastic gradient descent and backpropagation of errors, the model is trained for binary classification of translate...

BACKGROUND There is considerable international research regarding the link between human demographics and pet ownership. In several international studies, pet ownership was associated with household demographics including: the presence of children in the household, urban/rural location, level of education and age/family structure. What is lacking across all these studies, however, is an underst...

Given the low number of African Americans in Computer Science, we recognize the need to take effective measures to build and sustain a pipeline between K-12 education and institutions of higher learning with the hope that more African American students will pursue careers in computing related fields such as gaming. To support this effort, we engage African American middle school students in a 4...