Nonparametric Bayesian sparse factor models with application to gene expression modeling

A nonparametric Bayesian extension of Factor Analysis (FA) is proposed where observed data $\mathbf{Y}$ is modeled as a linear superposition, $\mathbf{G}$, of a potentially infinite number of hidden factors, $\mathbf{X}$. The Indian Buffet Process (IBP) is used as a prior on $\mathbf{G}$ to incorporate sparsity and to allow the number of latent features to be inferred. The model's utility for modeling gene expression data is investigated using randomly generated data sets based on a known sparse connectivity matrix for E. Coli, and on three biological data sets of increasing complexity.

Investigation of the losses in the stator duct spacers of large A.C. motors using the A -ip 3-D eddy current formulation

This paper describes how the A -if) formulation may be applied to determine the losses in the stator duct spacers of large a.c. motors. The model is described in terms of its geometry and boundary conditions. The novel aspects of the application of the formulation to this problem are explained. These include the modelling of fixed currents sources (the stator windings), the location of the necessary cut surfaces and the determination of their magnetic scalar potential differences, and the implementation of periodic boundary conditions for vector variables. Results are presented showing how the duct spacer losses vary with load, and with the relative permeability of the spacer material. The effects of modelling iron nonlinearity, of both the spacer and the steel laminations, are also illustrated. © 1996 IEEE.

Ballasted CNT structures for use as electron sources

The aim of this paper is to describe the growth and optimization of ballasted carbon nanotube (CNT) and CNT/Zinc Oxide nanostructures to produce novel electron sources for use in lighting and x-ray applications. © 2010 ITE and SID.