Learning the learning rate for prediction with expert advice

Most standard algorithms for prediction with expert advice depend on a parameter called the learning rate. This learning rate needs to be large enough to fit the data well, but small enough to prevent overfitting. For the exponential weights algorithm, a sequence of prior work has established theoretical guarantees for higher and higher data-dependent tunings of the learning rate, which allow for increasingly aggressive learning. But in practice such theoretical tunings often still perform worse (as measured by their regret) than ad hoc tuning with an even higher learning rate. To close the gap between theory and practice we introduce an approach to learn the learning rate. Up to a factor that is at most (poly)logarithmic in the number of experts and the inverse of the learning rate, our method performs as well as if we would know the empirically best learning rate from a large range that includes both conservative small values and values that are much higher than those for which formal guarantees were previously available. Our method employs a grid of learning rates, yet runs in linear time regardless of the size of the grid.

How everyday language can and will boost effective information retrieval

Typing 2 or 3 keywords into a browser has become an easy and efficient way to find information. Yet, typing even short queries becomes tedious on ever shrinking (virtual) keyboards. Meanwhile, speech processing is maturing rapidly, facilitating everyday language input. Also, wearable technology can inform users proactively by listening in on their conversations or processing their social media interactions. Given these developments, everyday language may soon become the new input of choice. We present an information retrieval (IR) algorithm specifically designed to accept everyday language. It integrates two paradigms of information retrieval, previously studied in isolation; one directed mainly at the surface structure of language, the other primarily at the underlying meaning. The integration was achieved by a Markov machine that encodes meaning by its transition graph, and surface structure by the language it generates. A rigorous evaluation of the approach showed, first, that it can compete with the quality of existing language models, second, that it is more effective the more verbose the input, and third, as a consequence, that it is promising for an imminent transition from keyword input, where the onus is on the user to formulate concise queries, to a modality where users can express more freely, more informal, and more natural their need for information in everyday language.

Exploiting inference from semantic annotations for information retrieval

The increasing amount of information that is annotated against standardised semantic resources offers opportunities to incorporate sophisticated levels of reasoning, or inference, into the retrieval process. In this position paper, we reflect on the need to incorporate semantic inference into retrieval (in particular for medical information retrieval) as well as previous attempts that have been made so far with mixed success. Medical information retrieval is a fertile ground for testing inference mechanisms to augment retrieval. The medical domain offers a plethora of carefully curated, structured, semantic resources, along with well established entity extraction and linking tools, and search topics that intuitively require a number of different inferential processes (e.g., conceptual similarity, conceptual implication, etc.). We argue that integrating semantic inference in information retrieval has the potential to uncover a large amount of information that otherwise would be inaccessible; but inference is also risky and, if not used cautiously, can harm retrieval.

Broadening the perspective on the airborne transmission of respiratory infection

Airborne bioaerosols are becoming increasingly recognized as a potential route of transmission for the spread of bacterial and viral respiratory tract infections.

A probabilistic framework for analysing the compositionality of conceptual combinations

Conceptual combination performs a fundamental role in creating the broad range of compound phrases utilised in everyday language. While the systematicity and productivity of language provide a strong argument in favour of assuming compositionality, this very assumption is still regularly questioned in both cognitive science and philosophy. This article provides a novel probabilistic framework for assessing whether the semantics of conceptual combinations are compositional, and so can be considered as a function of the semantics of the constituent concepts, or not. Rather than adjudicating between different grades of compositionality, the framework presented here contributes formal methods for determining a clear dividing line between compositional and non-compositional semantics. Compositionality is equated with a joint probability distribution modelling how the constituent concepts in the combination are interpreted. Marginal selectivity is emphasised as a pivotal probabilistic constraint for the application of the Bell/CH and CHSH systems of inequalities (referred to collectively as Bell-type). Non-compositionality is then equated with either a failure of marginal selectivity, or, in the presence of marginal selectivity, with a violation of Bell-type inequalities. In both non-compositional scenarios, the conceptual combination cannot be modelled using a joint probability distribution with variables corresponding to the interpretation of the individual concepts. The framework is demonstrated by applying it to an empirical scenario of twenty-four non-lexicalised conceptual combinations.

Quantum cognition: A new theoretical approach to psychology

What type of probability theory best describes the way humans make judgments under uncertainty and decisions under conflict? Although rational models of cognition have become prominent and have achieved much success, they adhere to the laws of classical probability theory despite the fact that human reasoning does not always conform to these laws. For this reason we have seen the recent emergence of models based on an alternative probabilistic framework drawn from quantum theory. These quantum models show promise in addressing cognitive phenomena that have proven recalcitrant to modeling by means of classical probability theory. This review compares and contrasts probabilistic models based on Bayesian or classical versus quantum principles, and highlights the advantages and disadvantages of each approach.

Impact of adolescent peer aggression on later educational and employment outcomes in an Australian cohort

This study used prospective birth cohort data to analyse the relationship between peer aggression at 14 years of age and educational and employment outcomes at 17 years (N = 1091) and 20 years (N = 1003). Participants from the Western Australian Pregnancy Cohort (Raine) study were divided into mutually exclusive categories of peer aggression. Involvement in peer aggression was reported by 40.2% (10.1% victims; 21.4% perpetrators; 8.7% victim–perpetrators) of participants. Participants involved in any form of peer aggression were less likely to complete secondary school. Perpetrators and victim–perpetrators of peer aggression were more likely to be in the ‘No Education, Employment or Training’ group at 20 years of age. This association was explained by non-completion of secondary school. These findings demonstrate a robust association between involvement in peer aggression and non-completion of secondary school, which in turn was associated with an increased risk of poor educational and employment outcomes in early adulthood.

Grassland productivity limited by multiple nutrients

Terrestrial ecosystem productivity is widely accepted to be nutrient limited1. Although nitrogen (N) is deemed a key determinant of aboveground net primary production (ANPP)2,3, the prevalence of co-limitation by N and phosphorus (P) is increasingly recognized4,​5,​6,​7,​8. However, the extent to which terrestrial productivity is co-limited by nutrients other than N and P has remained unclear. Here, we report results from a standardized factorial nutrient addition experiment, in which we added N, P and potassium (K) combined with a selection of micronutrients (K+μ), alone or in concert, to 42 grassland sites spanning five continents, and monitored ANPP. Nutrient availability limited productivity at 31 of the 42 grassland sites. And pairwise combinations of N, P, and K+μ co-limited ANPP at 29 of the sites. Nitrogen limitation peaked in cool, high latitude sites. Our findings highlight the importance of less studied nutrients, such as K and micronutrients, for grassland productivity, and point to significant variations in the type and degree of nutrient limitation. We suggest that multiple-nutrient constraints must be considered when assessing the ecosystem-scale consequences of nutrient enrichment.

Plant species’ origin predicts dominance and response to nutrient enrichment and herbivores in global grasslands

Exotic species dominate many communities; however the functional significance of species’ biogeographic origin remains highly contentious. This debate is fuelled in part by the lack of globally replicated, systematic data assessing the relationship between species provenance, function and response to perturbations. We examined the abundance of native and exotic plant species at 64 grasslands in 13 countries, and at a subset of the sites we experimentally tested native and exotic species responses to two fundamental drivers of invasion, mineral nutrient supplies and vertebrate herbivory. Exotic species are six times more likely to dominate communities than native species. Furthermore, while experimental nutrient addition increases the cover and richness of exotic species, nutrients decrease native diversity and cover. Native and exotic species also differ in their response to vertebrate consumer exclusion. These results suggest that species origin has functional significance, and that eutrophication will lead to increased exotic dominance in grasslands.

Anthropogenic nitrogen deposition predicts local grassland primary production worldwide

Humans dominate many important Earth system processes including the nitrogen (N) cycle. Atmospheric N deposition affects fundamental processes such as carbon cycling, climate regulation, and biodiversity, and could result in changes to fundamental Earth system processes such as primary production. Both modelling and experimentation have suggested a role for anthropogenically altered N deposition in increasing productivity, nevertheless, current understanding of the relative strength of N deposition with respect to other controls on production such as edaphic conditions and climate is limited. Here we use an international multiscale data set to show that atmospheric N deposition is positively correlated to aboveground net primary production (ANPP) observed at the 1-m2 level across a wide range of herbaceous ecosystems. N deposition was a better predictor than climatic drivers and local soil conditions, explaining 16% of observed variation in ANPP globally with an increase of 1 kg N·ha-1·yr-1 increasing ANPP by 3%. Soil pH explained 8% of observed variation in ANPP while climatic drivers showed no significant relationship. Our results illustrate that the incorporation of global N deposition patterns in Earth system models are likely to substantially improve estimates of primary production in herbaceous systems. In herbaceous systems across the world, humans appear to be partially driving local ANPP through impacts on the N cycle.

Genome-wide association analysis identifies 11 risk variants associated with the asthma with hay fever phenotype

Background To date, no genome-wide association study (GWAS) has considered the combined phenotype of asthma with hay fever. Previous analyses of family data from the Tasmanian Longitudinal Health Study provide evidence that this phenotype has a stronger genetic cause than asthma without hay fever. Objective We sought to perform a GWAS of asthma with hay fever to identify variants associated with having both diseases. Methods We performed a meta-analysis of GWASs comparing persons with both physician-diagnosed asthma and hay fever (n = 6,685) with persons with neither disease (n = 14,091). Results At genome-wide significance, we identified 11 independent variants associated with the risk of having asthma with hay fever, including 2 associations reaching this level of significance with allergic disease for the first time: ZBTB10 (rs7009110; odds ratio [OR], 1.14; P = 4 × 10−9) and CLEC16A (rs62026376; OR, 1.17; P = 1 × 10−8). The rs62026376:C allele associated with increased asthma with hay fever risk has been found to be associated also with decreased expression of the nearby DEXI gene in monocytes. The 11 variants were associated with the risk of asthma and hay fever separately, but the estimated associations with the individual phenotypes were weaker than with the combined asthma with hay fever phenotype. A variant near LRRC32 was a stronger risk factor for hay fever than for asthma, whereas the reverse was observed for variants in/near GSDMA and TSLP. Single nucleotide polymorphisms with suggestive evidence for association with asthma with hay fever risk included rs41295115 near IL2RA (OR, 1.28; P = 5 × 10−7) and rs76043829 in TNS1 (OR, 1.23; P = 2 × 10−6). Conclusion By focusing on the combined phenotype of asthma with hay fever, variants associated with the risk of allergic disease can be identified with greater efficiency.

Persistent organic pollutants in matched breast milk and infant faeces samples

Assessing blood concentration of persistent organic pollutants (POPs) in infants is difficult due to the ethical and practical difficulties in obtaining sufficient quantities of blood. To determine whether measuring POPs in faeces might reflect blood concentration during infancy, we measured the concentrations of a range of POPs (i.e. polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and organochlorine pesticides (OCPs)) in a pilot study using matched breast milk and infant faecal samples obtained from ten mother-child pairs. All infants were breast fed, with 8 of them also receiving solid food at the time of faecal sampling. In this small dataset faecal concentrations (range 0.01-41ngg-1 lipid) are strongly associated with milk concentrations (range 0.02-230ngg-1 lipid). Associations with other factors generally could not be detected in this dataset, with the exception of a small effect of age or growth. Different sources (external or internal) of exposure appeared to directly influence faecal concentrations of different chemicals based on different inter-individual variability in the faeces-to-milk concentration ratio Rfm. Overall, the matrix of faeces as an external measure of internal exposure in infants looks promising for some chemicals and is worth assessing further in larger datasets.

Isolation, purification, and full NMR assignments of cyclopamine from Veratrum californicum

The Hedgehog signaling pathway is essential for embryogenesis and for tissue homeostasis in the adult. However, it may induce malignancies in a number of tissues when constitutively activated, and it may also have a role in other forms of normal and maladaptive growth. Cyclopamine, a naturally occurring steroidal alkaloid, specifically inhibits the Hedgehog pathway by binding directly to Smoothened, an important Hedgehog response element. To use cyclopamine as a tool to explore and/or inhibit the Hedgehog pathway in vivo, a substantial quantity is required, and as a practical matter cyclopamine has been effectively unavailable for usage in animals larger than mice.

The coastal environment and human health: microbial indicators, pathogens, sentinels and reservoirs

Innovative research relating oceans and human health is advancing our understanding of disease-causing organisms in coastal ecosystems. Novel techniques are elucidating the loading, transport and fate of pathogens in coastal ecosystems, and identifying sources of contamination. This research is facilitating improved risk assessments for seafood consumers and those who use the oceans for recreation. A number of challenges still remain and define future directions of research and public policy. Sample processing and molecular detection techniques need to be advanced to allow rapid and specific identification of microbes of public health concern from complex environmental samples. Water quality standards need to be updated to more accurately reflect health risks and to provide managers with improved tools for decision-making. Greater discrimination of virulent versus harmless microbes is needed to identify environmental reservoirs of pathogens and factors leading to human infections. Investigations must include examination of microbial community dynamics that may be important from a human health perspective. Further research is needed to evaluate the ecology of non-enteric water-transmitted diseases. Sentinels should also be established and monitored, providing early warning of dangers to ecosystem health. Taken together, this effort will provide more reliable information about public health risks associated with beaches and seafood consumption, and how human activities can affect their exposure to disease-causing organisms from the oceans.