Deep generative models parameterized by neural networks have recently achieved state-ofthe-art performance in unsupervised and semisupervised learning. We extend deep generative models with auxiliary variables which improves the variational approximation. The auxiliary variables leave the generative model unchanged but make the variational distribution more expressive. Inspired by the structure...

We address the problem of learning good features for understanding video data. We introduce a model that learns latent representations of image sequences from pairs of successive images. The convolutional architecture of our model allows it to scale to realistic image sizes whilst using a compact parametrization. In experiments on the NORB dataset, we show our model extracts latent “flow fields...

The energy model is a simple, biologically inspired approach to extracting relationships between images in tasks like stereopsis and motion analysis. We discuss how adding an extra pooling layer to the energy model makes it possible to learn encodings of transformations that are mostly invariant with respect to image content, and to learn encodings of images that are mostly invariant with respe...

Motivated by the observation that coarse and fine resolutions of an image reveal different structures in the underlying visual phenomenon, we present a model based on the Deep Belief Network (DBN) which learns features from the multiscale representation of images. A Laplacian Pyramid is first constructed for each image. DBNs are then trained separately at each level of the pyramid. Finally, a t...

Recent advances in training deep (multi-layer) architectures have inspired a renaissance in neural network use. For example, deep convolutional networks are becoming the default option for difficult tasks on large datasets, such as image and speech recognition. However, here we show that error rates below 1% on the MNIST handwritten digit benchmark can be replicated with shallow non-convolution...

General purpose middleware has been shown effective in meeting diverse functional requirements for a wide range of distributed systems. Advanced middleware projects have also supported single quality-of-service dimensions such as real-time, fault tolerance, or small memory footprint. However, there is limited experience supporting multiple quality-of-service dimensions in middleware to meet the...

In classification tasks, restricted Boltzmann machines (RBMs) have predominantly been used in the first stage, either as feature extractors or to provide initialization of neural networks. In this study, we propose a discriminative learning approach to provide a self-contained RBM method for classification, inspired by free-energy based function approximation (FE-RBM), originally proposed for r...

Nitrous oxide (N2O) is not only recognized as a potent greenhouse gas, but it also used in industry clean energy source. In this study, different electrode materials of carbon felt and graphite were equipped the ΔnosZ P. aeruginosa/microbial electrolysis cell (MEC) systems to explore optimization mechanism for long-term N2O recovery during incineration leachate treatment. The group showed bette...

Efflux is an important resistance mechanism in Staphylococcus aureus, but its frequency in patients with bacteremia is unknown. Nonreplicate bloodstream isolates were collected over an 8-month period, and MICs of four common efflux pump substrates, with and without the broad-spectrum efflux pump inhibitor reserpine, were determined (n = 232). A reserpine-associated fourfold decrease in MIC was ...