The expected bit complexity of the von Neumann rejection algorithm

In 1952, von Neumann introduced the rejection method for random variate generation. We revisit this algorithm when we have a source of perfect bits at our disposal. In this random bit model, there are universal lower bounds for generating a random variate with a given density to within an accuracy \(\epsilon \) derived by Knuth and Yao, and refined by the authors. In general, von Neumann’s method fails in this model. We propose a modification that insures proper behavior for all Riemann-integrable densities on compact sets, and show that the expected number of random bits needed behaves optimally with respect to universal lower bounds. In particular, we introduce the notion of an oracle that evaluates the supremum and infimum of a function on any rectangle of \({\mathbb {R}}^{d}\), and develop a quadtree-style extension of the classical rejection method.     

Prior specification of neighbourhood and interaction structure in binary Markov random fields

We formulate a prior distribution for the energy function of stationary binary Markov random fields (MRFs) defined on a rectangular lattice. In the prior we assign distributions to all parts of the energy function. In particular we define priors for the neighbourhood structure of the MRF, what interactions to include in the model, and for potential values. We define a reversible jump Markov chain Monte Carlo (RJMCMC) procedure to simulate from the corresponding posterior distribution when conditioned to an observed scene. Thereby we are able to learn both the neighbourhood structure and the parametric form of the MRF from the observed scene. We circumvent evaluations of the intractable normalising constant of the MRF when running the RJMCMC algorithm by adopting a previously defined approximate auxiliary variable algorithm. We demonstrate the usefulness of our prior in two simulation examples and one real data example.     

Parallel and interacting stochastic approximation annealing algorithms for global optimisation

We present the parallel and interacting stochastic approximation annealing (PISAA) algorithm, a stochastic simulation procedure for global optimisation, that extends and improves the stochastic approximation annealing (SAA) by using population Monte Carlo ideas. The efficiency of standard SAA algorithm crucially depends on its self-adjusting mechanism which presents stability issues in high dimensional or rugged optimisation problems. The proposed algorithm involves simulating a population of SAA chains that interact each other in a manner that significantly improves the stability of the self-adjusting mechanism and the search for the global optimum in the sampling space, as well as it inherits SAA desired convergence properties when a square-root cooling schedule is used. It can be implemented in parallel computing environments in order to mitigate the computational overhead. As a result, PISAA can address complex optimisation problems that it would be difficult for SAA to satisfactory address. We demonstrate the good performance of the proposed algorithm on challenging applications including Bayesian network learning and protein folding. Our numerical comparisons suggest that PISAA outperforms the simulated annealing, stochastic approximation annealing, and annealing evolutionary stochastic approximation Monte Carlo.     

Is education a risk factor or social vaccine against HIV/AIDS in Sub-Saharan Africa? The effect of schooling across public health periods

Early in the 30-year HIV/AIDS pandemic in Sub-Saharan Africa, epidemiological studies identified formal education attainment as a risk factor: educated Sub-Saharan Africans had a higher risk of contracting HIV/AIDS than their less educated peers. Later demographic research reported that by the mid-1990s the education effect had reversed, and education began to function as a social vaccine. Recent counter-evidence finds a curvilinear pattern, with the association between educational attainment and HIV/AIDS infection changing from positive to negative across the education gradient. To reconcile these inconsistent conclusions, a hypothesis is developed and tested that education at early stages functioned as a risk factor and later functioned (and continues to function) as a social vaccine. We reason that this shift in the direction of the education effect was concurrent with changes in the public health environment in SSA that early on heightened material benefits from educational attainment but later heightened cognitive benefits from schooling. Using the 2003/2004 Demographic Health Surveys from four Sub-Saharan African countries (Cameroon, Ghana, Kenya and Tanzania), we tested this hypothesis (differential effects of schooling) using non-linear regression analysis (probit), identifying the different public health periods and controlling for confounding factors. The results support the hypothesis that the education effect shifted historically in the HIV/AIDS pandemic in SSA as we hypothesized.     

A Proposal for Public Sociology as Localized Intervention and Collective Enterprise: The Makings and Impact of <Emphasis Type="Italic">Invisible in Austin</Emphasis>

What can local public sociology look like, and what does it accomplish? This essay tracks the origins, makings and impacts of the book Invisible in Austin to evaluate its model of public sociology: as a collective enterprise with a local aim. Invisible in Austin: Life and Labor in an American City, the culmination of a three-year collaborative qualitative research project between a professor and twelve graduate students, depicts social suffering as lived for 11 individuals in Austin, Texas—a booming, highly segregated city with one of the country’s highest levels of income inequality. In its design, production, and effects, it envisions public sociology in a two-fold sense—in its joint, horizontal making, and in its intent to intervene in the local public sphere to make visible the daily lived experience of social marginality for those whose labor allows Austin to survive and thrive as a hip, creative technopolis—house cleaners, office machine repairers, cab drivers, restaurant cooks and dish washers, exotic dancers, musicians, and roofers, among them. Reflecting on the origins of the book, its joint assembling, and its outcomes thus far, we take stock of the lessons learned. In so doing, we provide a rubric for evaluating the wide spectrum of possible impacts of a public sociological intervention: through direct and indirect audience engagements, on the project’s subjects, and on local public policy. This reflection concludes with three suggestions: to approach public sociology as collective enterprise, to take narrative seriously, and to seek wide exposure.     

Individualism and Marriage: Ideal Types for Making Sense of the Relationship between Self and Sacrifice

The question of how the individual and group relate is one that has long interested social theorists. Changes in family form and structure in the contemporary West resituate this question in a contentious public debate regarding how the prevalence of new family forms may contribute or be deleterious to the well-being of individuals and families. Sociological discourse on marriage and the family generally tends to mirror this debate by dichotomizing individualism and commitment and self and marriage, resulting in an obfuscation of our understanding of the forms and styles in marriage. In order to clarify and advance this discussion, we show how individualism and commitment are mutually required in a modern world. We follow this by outlining a logically-derived typology that, along with a committed individualist and a group conformer, includes two intermediate types: a self-regulator and a relationship negotiator. We empirically demonstrate the utility of these types by showing how they correspond with the ways that interviewees talk about marriage in six local congregations, and we suggest various social factors that may particularly impact the development of local marriage cultures. These types provide a theoretical frame for understanding how individualism and commitment are intertwined and require each other.     

Gender and Performance Capital among Local Musicians

This article extends Bourdieu’s field theory to explain how learning spaces in the Toronto folk and metal scenes create gendered access to a field-specific form of cultural capital: performance capital, or the instrumental and interpersonal skills required to perform music. Folk musicians develop performance capital in open-access spaces, such as workshops and open stages at local folk clubs, while metal musicians learn in private spaces such as garages, basements, and rented rehearsal rooms. Folk’s learning spaces are open to all aspiring musicians, while access to heavy metal’s learning spaces relies on social networks from which women are often excluded. These different processes of capital development can lead to greater or lesser opportunities for women to become cultural producers: In Toronto, women make up approximately five percent of heavy metal musicians, but almost half of practicing folk musicians.