We consider a spatial generalized linear latent variable model with and without normality distributional assumption on the latent variables. When the latent variables are assumed to be multivariate normal, we apply a Laplace approximation. To relax the assumption of marginal normality in favor of a mixture of normals, we construct a multivariate density with Gaussian spatial dependence and give...

Latent variable models are used for analyzing multivariate data. Recently, generalized linear latent variable models for categorical, metric, and mixed-type responses estimated via maximum likelihood (ML) have been proposed. Model deviations, such as data contamination, are shown analytically, using the influence function and through a simulation study, to seriously affect ML estimation. This a...

Black box variational inference allows researchers to easily prototype and evaluate an array of models. Recent advances allow such algorithms to scale to high dimensions. However, a central question remains: How to specify an expressive variational distribution that maintains efficient computation? To address this, we develop hierarchical variational models (HVMs). HVMs augment a variational ap...

Unsupervised learning on imbalanced data is challenging because, when given imbalanced data, current model is often dominated by the major category and ignores the categories with small amount of data. We develop a latent variable model that can cope with imbalanced data by dividing the latent space into a shared space and a private space. Based on Gaussian Process Latent Variable Models, we pr...

Access to Higher Education and Inequality: The Chinese Experiment We apply a semi-parametric latent variable model to estimate selection and sorting effects on the evolution of private returns to schooling for college graduates during China’s reform between 1988 and 2002. We find that there were substantial sorting gains under the traditional system, but they have decreased drastically and beco...

In the medical, behavioral, social, and psychological sciences, latent variables represent unobservable traits that are measured by multiple observed variables. Latent variable models (LVMs) are useful tools for assessing the interrelationships among latent and observed variables. Due to their wide applications, LVMs have attracted significant attention from various fields. However, the majorit...

If there is one crucial problem with the application of so-called psychoJogical test theories in education (or in any other context). it is that both test deveJopers and test usersjust don 't know what they are doing. Let me give you two examples: First. we have been testing intelligence for three quarters ofa century now, but there are virtually no test psychologists who can tell what intellig...

Discriminative latent variable models (LVM) are frequently applied to various visual recognition tasks. In these systems the latent (hidden) variables provide a formalism for modeling structured variation of visual features. Conventionally, latent variables are defined on the variation of the foreground (positive) class. In this work we augment LVMs to include negative latent variables correspo...

We consider the problem of inference in a probabilistic model for transient populations where we wish to learn about arrivals, departures, and population size over all time, but the only available data are periodic counts of the population size at specific observation times. The underlying model arises in queueing theory (as an Mt/G/∞ queue) and also in ecological models for short-lived animals...

In this submission for MIREX 2014 we utilize an efficient latent variable model for multiple-F0 estimation and note tracking, which uses a variable-Q transform as a timefrequency representation. In contrast to the constant-Q transform of the 2013 submission, the variable-Q transform is able to provide a better temporal resolution in low frequencies with the same frequency resolution. The model ...