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 & Pregnancy. 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.

Recent advances in transition metal-based electrocatalysts for alkaline hydrogen evolution

Abstract: With the increasing demands in energy consumption and increasing environmental concerns, it is of vital significance for developing renewable and clean energy sources to substitute traditional fossil fuels. As an outstanding candidate, hydrogen is recognized as a green energy carrier due to its high gravimetric energy density, zero carbon footprints, and earth-abundance. Currently, water splitting in alkaline electrolytes represents one of the most promising methods for sustainable hydrogen production, and the key challenge lies in the development of high-performance electrocatalysts for the hydrogen evolution reaction (HER). Given the rapid advances in the design and development of efficient catalysts towards the alkaline HER, especially capable transition metal (TM)-based materials, this review aims to summarise recent progress in the theoretical understanding of the alkaline HER and TM-based electrocatalysts. TM-based catalysts classified by their different anionic compositions (metals, alloys, oxides, hydroxides, sulfides, selenides, tellurides, nitrides, phosphides, carbides, and borides) are comprehensively showcased. Special attention is given to mainstream strategies that can improve the catalytic properties of each category, as well as the underlying structure–activity regimes. Additionally, the challenges for the future development of novel catalysts are also analyzed.

Salivary melatonin as a circadian phase marker: validation and comparison to plasma melatonin

Abstract: There are many situations in which it would be useful to know the phase state of the biological clock. It is recognized that measurement of melatonin levels can provide this information, but traditionally blood has been used for the analysis, and there are many problems in extending the measurements into the home or field situations. The aim of this study was to develop and validate a salivary melatonin radioimmunoassay and to compare results obtained against a plasma assay for determining the onset of melatonin secretion. The assay developed was sensitive (4.3 pM) and required only 200 microliters of sample. A rhythm in melatonin was detected in saliva, peaking at approximately 120 pM or 30% of the plasma levels. Using an objective criterion for determining the onset of secretion (mean +/- 2 standard deviations of three daytime samples), the time of onset was shown to exhibit low intraindividual variability (coefficient of variation = 1.5%-4.3%). The time of onset determined using saliva was significantly correlated with the plasma onset (r = .70, p < .05). The onsets determined were 22:30 h +/- 22 min for the saliva and 21:50 h +/- 16 min for plasma for 17 subjects. Similarly, the acrophases of the saliva and plasma melatonin rhythms were significantly correlated. Neither posture alone nor changes in posture affected the calculation of the onset of melatonin secretion using the saliva approach. Very high saliva flow rates induced by citric acid resulted in lower melatonin concentrations compared to the gentle chewing on parafin film. These results firmly establish the use of salivary melatonin measurements for phase typing of the melatonin rhythm in humans.

BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation

Abstract: The low-level details and high-level semantics are both essential to the semantic segmentation task. However, to speed up the model inference, current approaches almost always sacrifice the low-level details, which leads to a considerable accuracy decrease. We propose to treat these spatial details and categorical semantics separately to achieve high accuracy and high efficiency for realtime semantic segmentation. To this end, we propose an efficient and effective architecture with a good trade-off between speed and accuracy, termed Bilateral Segmentation Network (BiSeNet V2). This architecture involves: (i) a Detail Branch, with wide channels and shallow layers to capture low-level details and generate high-resolution feature representation; (ii) a Semantic Branch, with narrow channels and deep layers to obtain high-level semantic context. The Semantic Branch is lightweight due to reducing the channel capacity and a fast-downsampling strategy. Furthermore, we design a Guided Aggregation Layer to enhance mutual connections and fuse both types of feature representation. Besides, a booster training strategy is designed to improve the segmentation performance without any extra inference cost. Extensive quantitative and qualitative evaluations demonstrate that the proposed architecture performs favourably against a few state-of-the-art real-time semantic segmentation approaches. Specifically, for a 2,048x1,024 input, we achieve 72.6% Mean IoU on the Cityscapes test set with a speed of 156 FPS on one NVIDIA GeForce GTX 1080 Ti card, which is significantly faster than existing methods, yet we achieve better segmentation accuracy.

Fast Supervised Hashing with Decision Trees for High-Dimensional Data

Abstract: Supervised hashing aims to map the original features to compact binary codes that are able to preserve label based similarity in the Hamming space. Non-linear hash functions have demonstrated their advantage over linear ones due to their powerful generalization capability. In the literature, kernel functions are typically used to achieve non-linearity in hashing, which achieve encouraging retrieval perfor- mance at the price of slow evaluation and training time. Here we propose to use boosted decision trees for achieving non-linearity in hashing, which are fast to train and evalu- ate, hence more suitable for hashing with high dimensional data. In our approach, we first propose sub-modular for- mulations for the hashing binary code inference problem and an efficient GraphCut based block search method for solving large-scale inference. Then we learn hash func- tions by training boosted decision trees to fit the binary codes. Experiments demonstrate that our proposed method significantly outperforms most state-of-the-art methods in retrieval precision and training time. Especially for high- dimensional data, our method is orders of magnitude faster than many methods in terms of training time.