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...

In the education industry, needs of online learning are significantly increasing. However, web-based courses demonstrate higher dropout rates than traditional courses. As a result, engaging students with data analysis is getting more crucial especially for distance learning. this study, we analyze on daily status in order to predict student’s schools. Specifically, trained prediction machine le...

BACKGROUND Medical school attrition is important--securing a place in medical school is difficult and a high attrition rate can affect the academic reputation of a medical school and staff morale. More important, however, are the personal consequences of dropout for the student. The aims of our study were to examine factors associated with attrition over a ten-year period (2001-2011) and to stu...