Australian National University

About: Australian National University is a education organization based out in Canberra, Australian Capital Territory, Australia. It is known for research contribution in the topics: Population & Galaxy. The organization has 34419 authors who have published 109261 publications receiving 4315448 citations. The organization is also known as: The Australian National University & ANU.

Caspase-11 cleaves gasdermin D for non-canonical inflammasome signalling

Abstract: Intracellular lipopolysaccharide from Gram-negative bacteria including Escherichia coli, Salmonella typhimurium, Shigella flexneri, and Burkholderia thailandensis activates mouse caspase-11, causing pyroptotic cell death, interleukin-1β processing, and lethal septic shock. How caspase-11 executes these downstream signalling events is largely unknown. Here we show that gasdermin D is essential for caspase-11-dependent pyroptosis and interleukin-1β maturation. A forward genetic screen with ethyl-N-nitrosourea-mutagenized mice links Gsdmd to the intracellular lipopolysaccharide response. Macrophages from Gsdmd(-/-) mice generated by gene targeting also exhibit defective pyroptosis and interleukin-1β secretion induced by cytoplasmic lipopolysaccharide or Gram-negative bacteria. In addition, Gsdmd(-/-) mice are protected from a lethal dose of lipopolysaccharide. Mechanistically, caspase-11 cleaves gasdermin D, and the resulting amino-terminal fragment promotes both pyroptosis and NLRP3-dependent activation of caspase-1 in a cell-intrinsic manner. Our data identify gasdermin D as a critical target of caspase-11 and a key mediator of the host response against Gram-negative bacteria.

The 2dF Galaxy Redshift Survey: spectra and redshifts

Abstract: The 2dF Galaxy Redshift Survey (2dFGRS) is designed to measure redshifts for approximately 250 000 galaxies. This paper describes the survey design, the spectroscopic observations, the redshift measurements and the survey data base. The 2dFGRS uses the 2dF multifibre spectrograph on the Anglo-Australian Telescope, which is capable of observing 400 objects simultaneously over a 2° diameter field. The source catalogue for the survey is a revised and extended version of the APM galaxy catalogue, and the targets are galaxies with extinction-corrected magnitudes brighter than b J = 19.45. The main survey regions are two declination strips, one in the southern Galactic hemisphere spanning 80° × 15° around the SGP, and the other in the northern Galactic hemisphere spanning 75° × 10° along the celestial equator; in addition, there are 99 fields spread over the southern Galactic cap. The survey covers 2000 deg 2 and has a median depth of z = 0.11. Adaptive tiling is used to give a highly uniform sampling rate of 93 per cent over the whole survey region. Redshifts are measured from spectra covering 3600-8000 A at a two-pixel resolution of 9.0 A and a median S/N of 13 pixel - 1 . All redshift identifications are visually checked and assigned a quality parameter Q in the range 1-5; Q ≥ 3 redshifts are 98.4 per cent reliable and have an rms uncertainty of 85 km s - 1 . The overall redshift completeness for Q ≥ 3 redshifts is 91.8 per cent, but this varies with magnitude from 99 per cent for the brightest galaxies to 90 per cent for objects at the survey limit. The 2dFGRS data base is available on the World Wide Web at http://www. mso.anu.edu.au/2dFGRS.

Boosting the margin: a new explanation for the effectiveness of voting methods

Abstract: One of the surprising recurring phenomena observed in experiments with boosting is that the test error of the generated classifier usually does not increase as its size becomes very large, and often is observed to decrease even after the training error reaches zero. In this paper, we show that this phenomenon is related to the distribution of margins of the training examples with respect to the generated voting classification rule, where the margin of an example is simply the difference between the number of correct votes and the maximum number of votes received by any incorrect label. We show that techniques used in the analysis of Vapnik's support vector classifiers and of neural networks with small weights can be applied to voting methods to relate the margin distribution to the test error. We also show theoretically and experimentally that boosting is especially effective at increasing the margins of the training examples. Finally, we compare our explanation to those based on the bias-variance decomposition.

Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering.

Abstract: Top-down visual attention mechanisms have been used extensively in image captioning and visual question answering (VQA) to enable deeper image understanding through fine-grained analysis and even multiple steps of reasoning. In this work, we propose a combined bottom-up and top-down attention mechanism that enables attention to be calculated at the level of objects and other salient image regions. This is the natural basis for attention to be considered. Within our approach, the bottom-up mechanism (based on Faster R-CNN) proposes image regions, each with an associated feature vector, while the top-down mechanism determines feature weightings. Applying this approach to image captioning, our results on the MSCOCO test server establish a new state-of-the-art for the task, achieving CIDEr / SPICE / BLEU-4 scores of 117.9, 21.5 and 36.9, respectively. Demonstrating the broad applicability of the method, applying the same approach to VQA we obtain first place in the 2017 VQA Challenge.