In this opinion piece we problematise foreign language (FL) learning curricula which are rigidly based on carefully sequenced input and constrained opportunities. Whilst an important component in learning, knowledge constitutes more than phonics vocabulary grammar. Our stance is premised the understanding that, whether novice or expert, FL learner not a ‘tabula rasa’. other words, every will ha...

INFERRING structured predictions based on correlated covariates remains a central problem in many fields, including NLP, computer vision, and computational biology. Typically, both the input covariates and output predictions can be high-dimensional, multi-modal, noisy, partially observable, and bearing latent structures, each of these characteristics adds a degree of complexity to the task of l...

In the refining industry, control room and field operators document their daily activities using shift logs. These logs are supposed to be an important part of the shift handover process and are the mechanism by which activities are coordinated across shifts. Previous research identified the need for a more structured approach to shift handover. However, the value of a structured approach has n...

| Using object recognition in simple scenes as the task, this research focuses on two fundamental problems in neural network systems: (1) processing large amounts of input with limited resources, and (2) the representation and use of structured knowledge. The rst problem arises because no practical neural network can process all the visual input simultaneously and e ciently. The solution is to ...

In this thesis we address structured machine learning problems. Here “structured” refers to situations in which the input or output domain of a prediction function is non-vectorial. Instead, the input instance or the predicted value can be decomposed into parts that follow certain dependencies, relations and constraints. Throughout the thesis we will use hard computer vision tasks as a rich sou...