Visual attention is the ability to select visual stimuli that are most behaviorally relevant among the many others. It allows us to allocate our limited processing resources to the most informative part of the visual scene. In this paper, we learn general high-level concepts with the aid of selective attention in a principled unsupervised framework, where a three layer deep network is built and...

Traditionally, when generative models of data are developed via deep architectures, greedy layer-wise pre-training is employed. In a well-trained model, the lower layer of the architecture models the data distribution conditional upon the hidden variables, while the higher layers model the hidden distribution prior. But due to the greedy scheme of the layerwise training technique, the parameter...

The paper deals with the modelling and analysis of thick piezoelectric multilayer composite shell continua applied to accurate and optimal design of adaptive (smart) structural components and systems in industrial applications. The smart composite is made of n laminae, in which each lamina can be used as either actuator, sensor, self-sensing actuator, or passive structural component. A discrete...

In this thesis, we present a method for learning problem-specific hierarchical features specialized for vision applications. Recently, a greedy layerwise learning mechanism has been proposed for tuning parameters of fully connected hierarchical networks. This approach views layers of a network as Restricted Boltzmann Machines (RBM), and trains them separately from the bottom layer upwards. We d...

In the Internet-age, malware poses a serious and evolving threat to security, making the detection of malware of utmost concern. Many research efforts have been conducted on intelligent malware detection by applying data mining and machine learning techniques. Though great results have been obtained with these methods, most of them are built on shallow learning architectures, which are still so...

BACKGROUND Glial cells are more numerous than neurons in the cortex and are crucial to neuronal function. There is evidence for reduced neuronal size in schizophrenia, with suggestive evidence for reduced glial cell density in mood disorders. In this investigation, we have simultaneously assessed glial cell density and neuronal density and size in the anterior cingulate cortex in schizophrenia,...

We introduce a novel training principle for probabilistic models that is an alternative to maximum likelihood. The proposed Generative Stochastic Networks (GSN) framework is based on learning the transition operator of a Markov chain whose stationary distribution estimates the data distribution. The transition distribution of the Markov chain is conditional on the previous state, generally invo...

A new three-noded thermomechanical beam finite element is derived for the analysis of laminated beams. The mechanical part is based on a refined model. The representation of the transverse shear strain by cosine function allows avoiding shear correction factors. This kinematics accounts for the interlaminar continuity conditions at the interfaces between the layers, and the boundary conditions ...

We introduce a novel training principle for generative probabilistic models that is an alternative to maximum likelihood. The proposed Generative Stochastic Networks (GSN) framework generalizes Denoising Auto-Encoders (DAE) and is based on learning the transition operator of a Markov chain whose stationary distribution estimates the data distribution. The transition distribution is a conditiona...

In recent years, convolutional neural networks (CNNs) have achieved remarkable success in various applications such as image classification, object detection, object parsing and face alignment. Such CNN models are extremely powerful to deal with massive amounts of training data by using millions and billions of parameters. However, these models are typically deficient due to the heavy cost in m...