In the present research, dynamic responses of multilayer functionally graded graphene platelets reinforced composite (FG-GPLRC) spherical panels under blast loading are studied. Three-dimensional elasticity theory is employed to derive governing equations. The distribution (GPLs) in each layer uniform and random with a constant weight fraction. GPLs concentration across panel thickness may be o...

In this study, an extended layerwise/solid-element (XLW/SE) method is developed for the thermo–chemo–mechanical (TCM) coupling problem of aero-engine turbine blade with thermal barrier coatings (TBCs). The consists two parts, layerwise (XLW) and three-dimensional (3D) solid-element (SE) method, which are adopted to formulate governing equations TBCs substrate, respectively. Then, according comp...

Adversarial training has been shown to regularize deep neural networks in addition to increasing their robustness to adversarial examples. However, its impact on very deep state of the art networks has not been fully investigated. In this paper, we present an efficient approach to perform adversarial training by perturbing intermediate layer activations and study the use of such perturbations a...

We use a simple density functional approach on a diffusional time scale, to address freezing to the body-centered cubic (bcc), hexagonal close-packed (hcp), and face-centered cubic (fcc) structures. We observe faceted equilibrium shapes and diffusion-controlled layerwise crystal growth consistent with two-dimensional nucleation. The predicted growth anisotropies are discussed in relation with r...

Learning multiple levels of feature detectors in Deep Belief Networks is a promising approach both for neuro-cognitive modeling and for practical applications, but it comes at the cost of high computational requirements. Here we propose a method for the parallelization of unsupervised generative learning in deep networks based on distributing training data among multiple computational nodes in ...

In this paper, we study two aspects of the variational autoencoder (VAE): the prior distribution over the latent variables and its corresponding posterior. First, we decompose the learning of VAEs into layerwise density estimation, and argue that having a flexible prior is beneficial to both sample generation and inference. Second, we analyze the family of inverse autoregressive flows (inverse ...

We studied the spin arrangement in ultrathin antiferromagnetic Mn films in contact with a ferromagnetic Fe(001) substrate using spin-polarized scanning tunneling microscopy. Mn shows a layerwise antiferromagnetic order on Fe(001). In regions where Mn overgrows Fe steps, a frustration of the antiferromagnetic order occurs which is similar to a 180 degrees domain wall. This topologically enforced...

Deep Boltzmann machines are in principle powerful models for extracting the hierarchical structure of data. Unfortunately, attempts to train layers jointly (without greedy layerwise pretraining) have been largely unsuccessful. We propose a modification of the learning algorithm that initially recenters the output of the activation functions to zero. This modification leads to a better condition...

In this work, we propose to train a deep neural network by distributed optimization over a graph. Two nonlinear functions are considered: the rectified linear unit (ReLU) and a linear unit with both lower and upper cutoffs (DCutLU). The problem reformulation over a graph is realized by explicitly representing ReLU or DCutLU using a set of slack variables. We then apply the alternating direction...