Based on generalized laminate plate theory, the formulation of a one-dimensional beam finite element with layer-wise constant shear (BLCS) is presented. The linear layer-wise representation of in-plane displacements permit accurate computation of normal stresses and transverse shear stresses on each layer for laminated beams with dissimilar ply stiffnesses. The BLCS formulation is equivalent to...

The bending behavior of composites sandwich plates with multi-layered laminated face sheets has been investigated, using a new four-nodded rectangular finite element formulation based on a layer-wise theory. Both, first order and higher-order shear deformation; theories are used in order to model the face sheets and the core, respectively. Unlike any other layer-wise theory, the number of degre...

In this paper, progressive damage and global failure of composite laminates under quasi-static, monotonic loading are investigated using 3D continuum damage mechanics. For this purpose, a finite element program has been developed using an eight-node 2D layered element including layer-wise plate theory. Damage analysis of a single orthotropic layer under various uniform in-plane and transverse l...

In this paper, we propose a deep convolutional neural network (CNN) with layer-wise context expansion and location-based attention, for large vocabulary speech recognition. In our model each higher layer uses information from broader contexts, along both the time and frequency dimensions, than its immediate lower layer. We show that both the layer-wise context expansion and the location-based a...

Deep Neural Networks (DNN) propose a new and efficient ML architecture based on the layer-wise building of several representation layers. A critical issue for DNNs remains model selection, e.g. selecting the number of neurons in each DNN layer. The hyper-parameter search space exponentially increases with the number of layers, making the popular grid search-based approach used for finding good ...

In Federated Learning (FL), a common approach for aggregating local solutions across clients is periodic full model averaging. It is, however, known that different layers of neural networks can have degree discrepancy the clients. The conventional aggregation scheme does not consider such difference and synchronizes whole parameters at once, resulting in inefficient network bandwidth consumptio...

Graph Convolutional Network (GCN) has been widely applied in transportation demand prediction due to its excellent ability capture non-Euclidean spatial dependence among station-level or regional demands. However, most of the existing research, graph convolution was implemented on a heuristically generated adjacency matrix, which could neither reflect real relationships stations accurately, nor...