Convolutional neural networks (convnets) have achieved impressive results on recent computer vision benchmarks. While they benefit from multiple layers that encode nonlinear decision boundaries and a degree of translation invariance, training convnets is a lengthy procedure fraught with local optima. Alternatively, a kernel method that incorporates the compositionality and symmetry of convnets ...

We present a new approximate inference algorithm for Deep Boltzmann Machines (DBM’s), a generative model with many layers of hidden variables. The algorithm learns a separate “recognition” model that is used to quickly initialize, in a single bottom-up pass, the values of the latent variables in all hidden layers. We show that using such a recognition model, followed by a combined top-down and ...

Convolutional Neural Networks have provided state-ofthe-art results in several computer vision problems. However, due to a large number of parameters in CNNs, they require a large number of training samples which is a limiting factor for small sample size problems. To address this limitation, we propose SSF-CNN which focuses on learning the “structure” and “strength” of filters. The structure o...

Convolutional neural networks (CNNs) have been successfully applied to many tasks such as digit and object recognition. Using convolutional (tied) weights significantly reduces the number of parameters that have to be learned, and also allows translational invariance to be hard-coded into the architecture. In this paper, we consider the problem of learning invariances, rather than relying on ha...

In this study, the addition of sulfamethazine (SMT) to landfill refuse decreased nitrogen intermediates (e.g. N2O and NO) and dinitrogen (N2) gas fluxes to <0.5 μg-N/kg-refuse·h-1, while the N2O and N2 flux were at ~1.5 and 5.0 μg-N/kg-refuse·h-1 respectively in samples to which oxytetracycline (OTC) had been added. The ARG (antibiotic resistance gene) levels in the refuse increased tenfold aft...

Expression of two genes of unknown function, Staphylococcus aureus scdA and Neisseria gonorrhoeae dnrN, is induced by exposure to oxidative or nitrosative stress. We show that DnrN and ScdA are di-iron proteins that protect their hosts from damage caused by exposure to nitric oxide and to hydrogen peroxide. Loss of FNR-dependent activation of aniA expression and NsrR-dependent repression of nor...

Manure composting is a significant source of atmospheric methane (CH4) and nitrous oxide (N2O) that are two potent greenhouse gases. The CH4 and N2O fluxes are mediated by methanogens and methanotrophs, nitrifying and denitrifying bacteria in composting manure, respectively, while these specific bacterial functional groups may interplay in CH4 and N2O emissions during manure composting. To test...

A gene cluster which includes genes required for the expression of nitric oxide reductase in Rhodobacter sphaeroides 2.4.3 has been isolated and characterized. Sequence analysis indicates that the two proximal genes in the cluster are the Nor structural genes. These two genes and four distal genes apparently constitute an operon. Mutational analysis indicates that the two structural genes, norC...

A nontoxigenic Aspergillus flavus strain, K49, is currently being tested as a biological control agent in corn fields in the Mississippi Delta. However, little is known about the overall genetic diversity of A. flavus from year to year in corn fields and specifically in Mississippi. Our objective was to assess the genetic variability of A. flavus isolates from different seasons, inoculum source...

The chief difficulty in object recognition is that objects’ classes are obscured by a large number of extraneous sources of variability, such as pose and part deformation. These sources of variation can be represented by symmetry groups, sets of composable transformations that preserve object identity. Convolutional neural networks (convnets) achieve a degree of translational invariance by comp...