We consider the parametric learning problem, where the objective of the learner is determined by a parametric loss function. Employing empirical risk minimization with possibly regularization, the inferred parameter vector will be biased toward the training samples. Such bias is measured by the cross validation procedure in practice where the data set is partitioned into a training set used for...

DISSECTING IMPLICIT LEADERSHIP THEORIES: A GENERALIZABILITYANALYSISby

The 2002 Lasker Clinical Medical Research Award honoured Willem Kolff and Belding Scribner, whose seminal work changed kidney failure from a fatal to a treatable disease. What began as exploratory efforts to sustain life and relieve uraemic symptoms in selected patients now provides chronic life-saving replacement therapy to millions of people worldwide. Continued improvements in clinical exper...

The problem of generalizability of empirical findings (experimental and observational) to new environments, settings, and populations is one of the central problems in causal inference. Experiments in the sciences are invariably conducted with the intent of being used elsewhere (e.g., outside the laboratory), where conditions are likely to be different. This practice is based on the premise tha...

of the Dissertation Generalizability in Causal Inference: Theory and Algorithms

Bug prediction is a technique used to estimate the most bug-prone entities in software systems. Bug prediction approaches vary in many design options, such as dependent variables, independent variables, and machine learning models. Choosing the right combination of design options to build an e↵ective bug predictor is hard. Previous studies do not consider this complexity and draw conclusions ba...

We have carried out a series of coverage evaluations of diverse types of parsers using texts from several genres such as newspaper, religious, legal and biomedical texts. We compared the overall coverage of the evaluated parsers and analyzed the differences by text genre. The results indicate that the coverage typically drops several percentage points when parsers are faced with texts on genres...