We propose a metric for evaluating the generalization ability of deep neural networks trained with mini-batch gradient descent. Our metric, called gradient disparity, is the l2 norm distance between the gradient vectors of two mini-batches drawn from the t ...
The past decades have seen numerous efforts to apply the finite volume methodology to solid mechanics problems. However, only limited work has been done by the finite volume community toward the simulation of mechanical contact. In this article, we present ...
We investigate the quasi-static growth of a fluid-driven frictional shear crack that propagates in mixed mode (II+III) on a planar fault interface that separates two identical half-spaces of a three-dimensional solid. The fault interface is characterized b ...
Dynamic non-contacting seals have been identified as the most appropriate sealing technology for reduced-scale and high-speed turbomachinery applications since they provide a clearance gap preventing rotor-to-seal contact during operation. Yet, any clearan ...
Deep neural networks have been empirically successful in a variety of tasks, however their theoretical understanding is still poor. In particular, modern deep neural networks have many more parameters than training data. Thus, in principle they should over ...
