BACKGROUND School dropout is a persisting problem with major socioeconomic consequences. Although poor health probably contributes to pathways leading to school dropout and health is likely negatively affected by dropout, these issues are relatively absent on the public health agenda. This emphasises the importance of integrative research aimed at identifying children at risk for school dropout...

Training deep belief networks (DBNs) requires optimizing a non-convex function with an extremely large number of parameters. Naturally, existing gradient descent (GD) based methods are prone to arbitrarily poor local minima. In this paper, we rigorously show that such local minima can be avoided (upto an approximation error) by using the dropout technique, a widely used heuristic in this domain...

BACKGROUND High school dropout and long-term sickness absence/disability pension in young adulthood are strongly associated. We investigated whether common risk factors in adolescence may confound this association. METHODS Data from 6612 school-attending adolescents (13-20 years old) participating in the Norwegian Young-HUNT1 Survey (1995-1997) was linked to long-term sickness absence or disa...

Regularisation of deep neural networks (DNN) during training is critical to performance. By far the most popular method is known as dropout. Here, cast through the prism of signal processing theory, we compare and contrast the regularisation effects of dropout with those of dither. We illustrate some serious inherent limitations of dropout and demonstrate that dither provides a far more effecti...

Between 20-51% of eating disorder clients in inpatient settings and 29-73% in outpatient settings prematurely discontinue treatment. Research has attempted to address this problem by identifying the client characteristics upon entering treatment that predict dropout, but this approach has failed to identify consistent predictors of treatment dropout. Processes that occur early in treatment, suc...

We investigate the use of alternative divergences to Kullback-Leibler (KL) in variational inference(VI), based on the Variational Dropout [10]. Stochastic gradient variational Bayes (SGVB) [9] is a general framework for estimating the evidence lower bound (ELBO) in Variational Bayes. In this work, we extend the SGVB estimator with using Alpha-Divergences, which are alternative to divergences to...

Limiting attrition (i.e., participant dropout before the conclusion of a study) is a major challenge faced by researchers when implementing clinical trials. Data from a smoking cessation trial for females (N = 246) were analyzed in order to identify baseline smoking-related, demographic and psychological characteristics affecting likelihood of early (i.e., before the quit attempt) and late (i.e...

Household finances are confidential and discussions limited to a subset of peers. We collect novel representative survey data examine separately whether interactions with inner outer social circles influence return perceptions, expectations, exposure widely known financial instrument in developed economy multiple information sources. find that respondent’s connectedness, proxied by perceived pr...