Training time on large datasets for deep neural networks is the principal workflow bottleneck in a number of important applications of deep learning, such as object classification and detection in automatic driver assistance systems (ADAS). To minimize training time, the training of a deep neural network must be scaled beyond a single machine to as many machines as possible by distributing the ...

The Saccharomyces Genome Database (SGD) is the main community repository of information for the budding yeast, Saccharomyces cerevisiae. The SGD has collected published results on chromosomal features, including genes and their products, and has become an encyclopedia of information on the biology of the yeast cell. This information includes gene and gene product function, phenotype, interactio...

The Saccharomyces Genome Database (SGD) provides Internet access to the complete Saccharomyces cerevisiae genomic sequence, its genes and their products, the phenotypes of its mutants, and the literature supporting these data. The amount of information and the number of features provided by SGD have increased greatly following the release of the S.cerevisiae genomic sequence, which is currently...

We constructed a survey system of radon/methane/nitrate/salinity to find sites of submarine groundwater discharge (SGD) and groundwater nitrate input. We deployed the system in Waquoit Bay and Boston Harbor, MA where we derived SGD rates using a mass balance of radon with methane serving as a fine resolution qualitative indicator of groundwater. In Waquoit Bay we identified several locations of...

Neutrophil specific granule deficiency (SGD) is a congenital disorder associated with an impaired inflammatory response and a deficiency of several granule proteins. The underlying abnormality causing the deficiencies is unknown. We examined mRNA transcription and protein synthesis of two neutrophil granule proteins, lactoferrin and myeloperoxidase in SGD. Metabolically labeled SGD nucleated ma...

Stochastic Gradient Descent (SGD) is an important algorithm in machine learning. With constant learning rates, it is a stochastic process that, after an initial phase of convergence, generates samples from a stationary distribution. We show that SGD with constant rates can be effectively used as an approximate posterior inference algorithm for probabilistic modeling. Specifically, we show how t...

The exchange of groundwater between land and sea is a major component of the hydrological cycle. This exchange, called submarine groundwater discharge (SGD), is comprised of terrestrial water mixed with sea water that has infiltrated coastal aquifers. The composition of SGD differs from that predicted by simple mixing because biogeochemical reactions in the aquifer modify its chemistry. To emph...

The Saccharomyces Genome Database (SGD; http://www.yeastgenome.org) is an expertly curated database of literature-derived functional information for the model organism budding yeast, Saccharomyces cerevisiae. SGD constantly strives to synergize new types of experimental data and bioinformatics predictions with existing data, and to organize them into a comprehensive and up-to-date information r...

Mini-batch based Stochastic Gradient Descent(SGD) has been widely used to train deep neural networks efficiently. In this paper, we design a general framework to automatically and adaptively select training data for SGD. The framework is based on neural networks and we call it Neural Data Filter (NDF). In Neural Data Filter, the whole training process of the original neural network is monitored...