University of Adelaide

About: University of Adelaide is a education organization based out in Adelaide, South Australia, Australia. It is known for research contribution in the topics: Population & Poison control. The organization has 27251 authors who have published 79167 publications receiving 2671128 citations. The organization is also known as: The University of Adelaide & Adelaide University.

What Value Do Explicit High Level Concepts Have in Vision to Language Problems

Abstract: Much recent progress in Vision-to-Language (V2L) problems has been achieved through a combination of Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). This approach does not explicitly represent high-level semantic concepts, but rather seeks to progress directly from image features to text. In this paper we investigate whether this direct approach succeeds due to, or despite, the fact that it avoids the explicit representation of high-level information. We propose a method of incorporating high-level concepts into the successful CNN-RNN approach, and show that it achieves a significant improvement on the state-of-the-art in both image captioning and visual question answering. We also show that the same mechanism can be used to introduce external semantic information and that doing so further improves performance. We achieve the best reported results on both image captioning and VQA on several benchmark datasets, and provide an analysis of the value of explicit high-level concepts in V2L problems.

Extending an evidence hierarchy to include topics other than treatment: revising the Australian 'levels of evidence'

Abstract: In 1999 a four-level hierarchy of evidence was promoted by the National Health and Medical Research Council in Australia. The primary purpose of this hierarchy was to assist with clinical practice guideline development, although it was co-opted for use in systematic literature reviews and health technology assessments. In this hierarchy interventional study designs were ranked according to the likelihood that bias had been eliminated and thus it was not ideal to assess studies that addressed other types of clinical questions. This paper reports on the revision and extension of this evidence hierarchy to enable broader use within existing evidence assessment systems. A working party identified and assessed empirical evidence, and used a commissioned review of existing evidence assessment schema, to support decision-making regarding revision of the hierarchy. The aim was to retain the existing evidence levels I-IV but increase their relevance for assessing the quality of individual diagnostic accuracy, prognostic, aetiologic and screening studies. Comprehensive public consultation was undertaken and the revised hierarchy was piloted by individual health technology assessment agencies and clinical practice guideline developers. After two and a half years, the hierarchy was again revised and commenced a further 18 month pilot period. A suitable framework was identified upon which to model the revision. Consistency was maintained in the hierarchy of "levels of evidence" across all types of clinical questions; empirical evidence was used to support the relationship between study design and ranking in the hierarchy wherever possible; and systematic reviews of lower level studies were themselves ascribed a ranking. The impact of ethics on the hierarchy of study designs was acknowledged in the framework, along with a consideration of how harms should be assessed. The revised evidence hierarchy is now widely used and provides a common standard against which to initially judge the likelihood of bias in individual studies evaluating interventional, diagnostic accuracy, prognostic, aetiologic or screening topics. Detailed quality appraisal of these individual studies, as well as grading of the body of evidence to answer each clinical, research or policy question, can then be undertaken as required.

Approaches for measuring the surface areas of metal oxide electrocatalysts for determining their intrinsic electrocatalytic activity

Abstract: Great attention has been recently drawn to metal oxide electrocatalysts for electrocatalysis-based energy storage and conversion devices. To find the optimal electrocatalyst, a prerequisite is an activity metric that reasonably evaluates the intrinsic electrocatalytic activity of a particular catalyst. The intrinsic activity is commonly defined as the specific activity which is the current per unit catalyst surface area. Thus, the precise assessment of intrinsic activity highly depends on the reliable measurement of catalyst surface area, which calls for the knowledge of experimental approaches for determining the surface areas of metal oxide electrocatalysts. This tutorial review aims to summarize and analyze the approaches for measuring the surface areas of metal oxide electrocatalysts for evaluating and comparing their intrinsic electrocatalytic activities. We start by comparing the popular metrics for activity estimation and highlighting the importance of surface-area-normalized activity (i.e. specific activity) for intrinsic chemistry analysis. Second, we provide some general guidelines for experimentally measuring the electrochemically active surface area (ECSA). Third, we review the methods for the surface area measurement of metal oxide electrocatalysts. The detailed procedure for each method is explicitly described to provide a step-by-step manual that guides researchers to perform the measurement; the rationales and uncertainties for each method are discussed to help readers justify the reliable assessment of surface area. Next, we give our recommendations on selecting a rational experimental approach for the surface area measurement of a particular metal oxide electrocatalyst. Lastly, we discuss the future challenges of ECSA measurement and present an exemplary novel ECSA technique.