Gradient factor centering is a new methodology for decomposing neural networks into biased and centered subnets which are then trained in parallel. The decomposition can be applied to any pattern-dependent factor in the network’s gradient, and is designed such that the subnets are more amenable to optimization by gradient descent than the original network: biased subnets because of their simpli...

We describe in this paper an automated method for assessing local coherence in short argumentative essays. We use ideas from Centering Theory to measure local coherence of essays’ paragraphs and compare it to human judgments on one analytical feature of essay quality called Continuity. Paragraphs which correspond to a discourse segment in our work and which are dominated by one prominent concep...

Abstract In this research, the feasibility of strengthening self-centering walls by high-performance concrete was investigated through an experimentally validated finite element model. The effects wall’s axial stress and tendons’ prestressing ratios on damage, effectiveness potential approaches were studied 40 independent scenarios, 360 different cases. Using estimated damage from numerical res...

Hybrid Coupled Wall (HCW) systems consist of reinforced concrete walls connected with steel coupling beams. HCWs benefit from the superior lateral stiffness walls, while mechanism reduces moment demand at base walls. The present study investigates seismic performance a new HCW system equipped friction-damped self-centering beams and examines efficiency in reducing residual deformations. intende...

The radial array of cytoplasmic microtubules (MTs) provides routes for intracellular transport and defines spatial organization of cytoplasm through interaction with molecular motors bound to membrane organelles (Kellogg et al., 1994; Hirokawa, 1998). The array is believed to be organized by the centrosome, which is capable of nucleating MTs. Our recent studies of pigment transport (Rodionov an...