Abstract: Omega-3 polyunsaturated fatty acids (PUFAs) include α-linolenic acid (ALA; 18:3 ω-3), stearidonic acid (SDA; 18:4 ω-3), eicosapentaenoic acid (EPA; 20:5 ω-3), docosapentaenoic acid (DPA; 22:5 ω-3), and docosahexaenoic acid (DHA; 22:6 ω-3). In the past few decades, many epidemiological studies have been conducted on the myriad health benefits of omega-3 PUFAs. In this review, we summarized the structural features, properties, dietary sources, metabolism, and bioavailability of omega-3 PUFAs and their effects on cardiovascular disease, diabetes, cancer, Alzheimer's disease, dementia, depression, visual and neurological development, and maternal and child health. Even though many health benefits of omega-3 PUFAs have been reported in the literature, there are also some controversies about their efficacy and certain benefits to human health.

Abstract: The kagome lattice is a two-dimensional network of corner-sharing triangles that is known to host exotic quantum magnetic states. Theoretical work has predicted that kagome lattices may also host Dirac electronic states that could lead to topological and Chern insulating phases, but these states have so far not been detected in experiments. Here we study the d-electron kagome metal Fe3Sn2, which is designed to support bulk massive Dirac fermions in the presence of ferromagnetic order. We observe a temperature-independent intrinsic anomalous Hall conductivity that persists above room temperature, which is suggestive of prominent Berry curvature from the time-reversal-symmetry-breaking electronic bands of the kagome plane. Using angle-resolved photoemission spectroscopy, we observe a pair of quasi-two-dimensional Dirac cones near the Fermi level with a mass gap of 30 millielectronvolts, which correspond to massive Dirac fermions that generate Berry-curvature-induced Hall conductivity. We show that this behaviour is a consequence of the underlying symmetry properties of the bilayer kagome lattice in the ferromagnetic state and the atomic spin-orbit coupling. This work provides evidence for a ferromagnetic kagome metal and an example of emergent topological electronic properties in a correlated electron system. Our results provide insight into the recent discoveries of exotic electronic behaviour in kagome-lattice antiferromagnets and may enable lattice-model realizations of fractional topological quantum states.

Abstract: Convolutional Neural Networks (CNN) are state-of-theart models for many image classification tasks. However, to recognize cancer subtypes automatically, training a CNN on gigapixel resolution Whole Slide Tissue Images (WSI) is currently computationally impossible. The differentiation of cancer subtypes is based on cellular-level visual features observed on image patch scale. Therefore, we argue that in this situation, training a patch-level classifier on image patches will perform better than or similar to an image-level classifier. The challenge becomes how to intelligently combine patch-level classification results and model the fact that not all patches will be discriminative. We propose to train a decision fusion model to aggregate patch-level predictions given by patch-level CNNs, which to the best of our knowledge has not been shown before. Furthermore, we formulate a novel Expectation-Maximization (EM) based method that automatically locates discriminative patches robustly by utilizing the spatial relationships of patches. We apply our method to the classification of glioma and non-small-cell lung carcinoma cases into subtypes. The classification accuracy of our method is similar to the inter-observer agreement between pathologists. Although it is impossible to train CNNs on WSIs, we experimentally demonstrate using a comparable non-cancer dataset of smaller images that a patch-based CNN can outperform an image-based CNN.

Abstract: We revisit the one-shot Neural Architecture Search (NAS) paradigm and analyze its advantages over existing NAS approaches. Existing one-shot method, however, is hard to train and not yet effective on large scale datasets like ImageNet. This work propose a Single Path One-Shot model to address the challenge in the training. Our central idea is to construct a simplified supernet, where all architectures are single paths so that weight co-adaption problem is alleviated. Training is performed by uniform path sampling. All architectures (and their weights) are trained fully and equally.

Abstract: We present a variational approximation to the information bottleneck of Tishby et al. (1999). This variational approach allows us to parameterize the information bottleneck model using a neural network and leverage the reparameterization trick for efficient training. We call this method "Deep Variational Information Bottleneck", or Deep VIB. We show that models trained with the VIB objective outperform those that are trained with other forms of regularization, in terms of generalization performance and robustness to adversarial attack.

Abstract: Feature selection (FS) methods can be used in data pre-processing to achieve efficient data reduction. This is useful for finding accurate data models. Since exhaustive search for optimal feature subset is infeasible in most cases, many search strategies have been proposed in literature. The usual applications of FS are in classification, clustering, and regression tasks. This review considers most of the commonly used FS techniques. Particular emphasis is on the application aspects. In addition to standard filter, wrapper, and embedded methods, we also provide insight into FS for recent hybrid approaches and other advanced topics.

Abstract: We present our experience with QUIC, an encrypted, multiplexed, and low-latency transport protocol designed from the ground up to improve transport performance for HTTPS traffic and to enable rapid deployment and continued evolution of transport mechanisms. QUIC has been globally deployed at Google on thousands of servers and is used to serve traffic to a range of clients including a widely-used web browser (Chrome) and a popular mobile video streaming app (YouTube). We estimate that 7% of Internet traffic is now QUIC. We describe our motivations for developing a new transport, the principles that guided our design, the Internet-scale process that we used to perform iterative experiments on QUIC, performance improvements seen by our various services, and our experience deploying QUIC globally. We also share lessons about transport design and the Internet ecosystem that we learned from our deployment.

Abstract: It is estimated that the world population will reach a record 7.3 billion in 2015, and the high burden of chronic conditions associated with ageing and smoking will increase further. Respiratory diseases in general receive little attention and funding in comparison with other major causes of global morbidity and mortality. In particular, chronic obstructive pulmonary disease (COPD) has been a major public health problem and will remain a challenge for clinicians within the 21st century. Worldwide, COPD is in the spotlight, since its high prevalence, morbidity and mortality create formidable challenges for health-care systems. This review emphasizes the magnitude of the COPD problem from a clinician's standpoint by drawing extensively from the new findings of the Global Burden of Disease study. Updated, distilled information on the population distribution of COPD is useful for the clinician to help provide an appreciation of the relative impact of COPD in daily practice compared with other chronic conditions, and to allocate minimum resources in anticipation of future needs in care. Despite recent trends in reduction of COPD standardized mortality rates and some recent successes in anti-smoking efforts in a number of Western countries, the overarching demographic impact of ageing in an ever-expanding world population, joined with other factors such as high rates of smoking and air pollution in Asia, will ensure that COPD will continue to pose an ever-increasing problem well into the 21st century.

Abstract: This review describes recent developments related to the unified model of active galactic nuclei (AGNs). It focuses on new ideas about the origin and properties of the central obscurer (torus) and the connection to its surroundings. The review does not address radio unification. AGN tori must be clumpy but uncertainties about their properties persist. Today's most promising models involve disk winds of various types and hydrodynamic simulations that link the large-scale galactic disk to the inner accretion flow. Infrared (IR) studies greatly improved our understanding of the spectral energy distribution of AGNs, but they are hindered by various selection effects. X-ray samples are more complete. The dependence of the covering factor of the torus on luminosity is a basic relationship that remains unexplained. There is also much confusion regarding real type-II AGNs, which do not fit into a simple unification scheme. The most impressive recent results are due to IR interferometry, which is not in accord wit...

Abstract: The in silico fragmenter MetFrag, launched in 2010, was one of the first approaches combining compound database searching and fragmentation prediction for small molecule identification from tandem mass spectrometry data. Since then many new approaches have evolved, as has MetFrag itself. This article details the latest developments to MetFrag and its use in small molecule identification since the original publication. MetFrag has gone through algorithmic and scoring refinements. New features include the retrieval of reference, data source and patent information via ChemSpider and PubChem web services, as well as InChIKey filtering to reduce candidate redundancy due to stereoisomerism. Candidates can be filtered or scored differently based on criteria like occurence of certain elements and/or substructures prior to fragmentation, or presence in so-called “suspect lists”. Retention time information can now be calculated either within MetFrag with a sufficient amount of user-provided retention times, or incorporated separately as “user-defined scores” to be included in candidate ranking. The changes to MetFrag were evaluated on the original dataset as well as a dataset of 473 merged high resolution tandem mass spectra (HR-MS/MS) and compared with another open source in silico fragmenter, CFM-ID. Using HR-MS/MS information only, MetFrag2.2 and CFM-ID had 30 and 43 Top 1 ranks, respectively, using PubChem as a database. Including reference and retention information in MetFrag2.2 improved this to 420 and 336 Top 1 ranks with ChemSpider and PubChem (89 and 71 %), respectively, and even up to 343 Top 1 ranks (PubChem) when combining with CFM-ID. The optimal parameters and weights were verified using three additional datasets of 824 merged HR-MS/MS spectra in total. Further examples are given to demonstrate flexibility of the enhanced features. In many cases additional information is available from the experimental context to add to small molecule identification, which is especially useful where the mass spectrum alone is not sufficient for candidate selection from a large number of candidates. The results achieved with MetFrag2.2 clearly show the benefit of considering this additional information. The new functions greatly enhance the chance of identification success and have been incorporated into a command line interface in a flexible way designed to be integrated into high throughput workflows. Feedback on the command line version of MetFrag2.2 available at http://c-ruttkies.github.io/MetFrag/ is welcome.

Abstract: Importance Oxidative stress and vascular impairment are believed to partly mediate age-related cognitive decline, a strong risk factor for development of dementia. Epidemiologic studies suggest that a Mediterranean diet, an antioxidant-rich cardioprotective dietary pattern, delays cognitive decline, but clinical trial evidence is lacking. Objective To investigate whether a Mediterranean diet supplemented with antioxidant-rich foods influences cognitive function compared with a control diet. Design, Setting, and Participants Parallel-group randomized clinical trial of 447 cognitively healthy volunteers from Barcelona, Spain (233 women [52.1%]; mean age, 66.9 years), at high cardiovascular risk were enrolled into the Prevencion con Dieta Mediterranea nutrition intervention trial from October 1, 2003, through December 31, 2009. All patients underwent neuropsychological assessment at inclusion and were offered retesting at the end of the study. Interventions Participants were randomly assigned to a Mediterranean diet supplemented with extravirgin olive oil (1 L/wk), a Mediterranean diet supplemented with mixed nuts (30 g/d), or a control diet (advice to reduce dietary fat). Main Outcomes and Measures Rates of cognitive change over time based on a neuropsychological test battery: Mini-Mental State Examination, Rey Auditory Verbal Learning Test (RAVLT), Animals Semantic Fluency, Digit Span subtest from the Wechsler Adult Intelligence Scale, Verbal Paired Associates from the Wechsler Memory Scale, and the Color Trail Test. We used mean z scores of change in each test to construct 3 cognitive composites: memory, frontal (attention and executive function), and global. Results Follow-up cognitive tests were available in 334 participants after intervention (median, 4.1 years). In multivariate analyses adjusted for confounders, participants allocated to a Mediterranean diet plus olive oil scored better on the RAVLT ( P = .049) and Color Trail Test part 2 ( P = .04) compared with controls; no between-group differences were observed for the other cognitive tests. Similarly adjusted cognitive composites (mean z scores with 95% CIs) for changes above baseline of the memory composite were 0.04 (−0.09 to 0.18) for the Mediterranean diet plus olive oil, 0.09 (−0.05 to 0.23; P = .04 vs controls) for the Mediterranean diet plus nuts, and −0.17 (−0.32 to −0.01) for the control diet. Respective changes from baseline of the frontal cognition composite were 0.23 (0.03 to 0.43; P = .003 vs controls), 0.03 (−0.25 to 0.31), and −0.33 (−0.57 to −0.09). Changes from baseline of the global cognition composite were 0.05 (−0.11 to 0.21; P = .005 vs controls) for the Mediterranean diet plus olive oil, −0.05 (−0.27 to 0.18) for the Mediterranean diet plus nuts, and −0.38 (−0.57 to −0.18) for the control diet. All cognitive composites significantly ( P Conclusions and Relevance In an older population, a Mediterranean diet supplemented with olive oil or nuts is associated with improved cognitive function. Trial Registration isrctn.org Identifier:ISRCTN35739639

Abstract: Social media allows people to freely interact with others and offers multiple ways for marketers to reach and engage with consumers. Considering the numerous ways social media affects individuals and businesses alike, in this article, the authors focus on where they believe the future of social media lies when considering marketing-related topics and issues. Drawing on academic research, discussions with industry leaders, and popular discourse, the authors identify nine themes, organized by predicted imminence (i.e., the immediate, near, and far futures), that they believe will meaningfully shape the future of social media through three lenses: consumer, industry, and public policy. Within each theme, the authors describe the digital landscape, present and discuss their predictions, and identify relevant future research directions for academics and practitioners.

Abstract: Objective: To report two patients infected with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) who acutely presented with Miller Fisher syndrome and polyneuritis cranialis, respectively. Methods: Patient data were obtained from medical records from the University Hospital “Principe de Asturias”, Alcala de Henares, Madrid, Spain and from the University Hospital “12 de Octubre”, Madrid, Spain. Results: The first patient was a 50-year-old man who presented with anosmia, ageusia, right internuclear ophthalmoparesis, right fascicular oculomotor palsy, ataxia, areflexia, albuminocytologic dissociation and positive testing for GD1b-IgG antibodies. Five days before, he had developed a cough, malaise, headache, low back pain, and a fever. The second patient was a 39-year-old man who presented with ageusia, bilateral abducens palsy, areflexia and albuminocytologic dissociation. Three days before, he had developed diarrhea, a low-grade fever, and a poor general condition. The oropharyngeal swab test for coronavirus disease 2019 (COVID-19) by qualitative real-time reverse-transcriptase–polymerase-chain-reaction assay was positive in both patients and negative in the cerebrospinal fluid. The first patient was treated with intravenous immunoglobulin and the second, with acetaminophen. Two weeks later, both patients made a complete neurological recovery, except for residual anosmia and ageusia in the first case. Conclusions: Our two cases highlight the rare occurrence of Miller Fisher syndrome and polyneuritis cranialis during the COVID-2 pandemic. Neurological manifestations may occur because of an aberrant immune response to COVID-19. The full clinical spectrum of neurological symptoms in patients with COVID-19 remains to be characterized.

Abstract: Importance Glioblastoma multiforme (GBM) remains almost invariably fatal despite optimal surgical and medical therapy. The association between the extent of tumor resection (EOR) and outcome remains undefined, notwithstanding many relevant studies. Objective To determine whether greater EOR is associated with improved 1- and 2-year overall survival and 6-month and 1-year progression-free survival in patients with GBM. Data Sources Pubmed, CINAHL, and Web of Science (January 1, 1966, to December 1, 2015) were systematically reviewed with librarian guidance. Additional articles were included after consultation with experts and evaluation of bibliographies. Articles were collected from January 15 to December 1, 2015. Study Selection Studies of adult patients with newly diagnosed supratentorial GBM comparing various EOR and presenting objective overall or progression-free survival data were included. Pediatric studies were excluded. Data Extraction and Synthesis Data were extracted from the text of articles or the Kaplan-Meier curves independently by investigators who were blinded to each other’s results. Data were analyzed to assess mortality after gross total resection (GTR), subtotal resection (STR), and biopsy. The body of evidence was evaluated according to Grading of Recommendations Assessment, Development, and Evaluation (GRADE) criteria and PRISMA guidelines. Main Outcome and Measures Relative risk (RR) for mortality at 1 and 2 years and progression at 6 months and 1 year. Results The search produced 37 studies suitable for inclusion (41 117 unique patients). The meta-analysis revealed decreased mortality for GTR compared with STR at 1 year (RR, 0.62; 95% CI, 0.56-0.69; P P P P P = .04; NNT, 593). The likelihood of disease progression was decreased with GTR compared with STR at 6 months (RR, 0.72; 95% CI, 0.48-1.09; P = .12; NNT, 14) and 1 year (RR, 0.66; 95% CI, 0.43-0.99; P Conclusion and Relevance This analysis represents the largest systematic review and only quantitative systematic review to date performed on this subject. Compared with STR, GTR substantially improves overall and progression-free survival, but the quality of the supporting evidence is moderate to low.

Abstract: We apply basic statistical reasoning to signal reconstruction by machine learning -- learning to map corrupted observations to clean signals -- with a simple and powerful conclusion: it is possible to learn to restore images by only looking at corrupted examples, at performance at and sometimes exceeding training using clean data, without explicit image priors or likelihood models of the corruption. In practice, we show that a single model learns photographic noise removal, denoising synthetic Monte Carlo images, and reconstruction of undersampled MRI scans -- all corrupted by different processes -- based on noisy data only.

Abstract: Worldwide glacier retreat and associated future runoff changes raise major concerns over the sustainability of global water resources1–4, but global-scale assessments of glacier decline and the resulting hydrological consequences are scarce5,6. Here we compute global glacier runoff changes for 56 large-scale glacierized drainage basins to 2100 and analyse the glacial impact on streamflow. In roughly half of the investigated basins, the modelled annual glacier runoff continues to rise until a maximum (‘peak water’) is reached, beyond which runoff steadily declines. In the remaining basins, this tipping point has already been passed. Peak water occurs later in basins with larger glaciers and higher ice-cover fractions. Typically, future glacier runoff increases in early summer but decreases in late summer. Although most of the 56 basins have less than 2% ice coverage, by 2100 one-third of them might experience runoff decreases greater than 10% due to glacier mass loss in at least one month of the melt season, with the largest reductions in central Asia and the Andes. We conclude that, even in large-scale basins with minimal ice-cover fraction, the downstream hydrological effects of continued glacier wastage can be substantial, but the magnitudes vary greatly among basins and throughout the melt season.

Abstract: In the current age of the Fourth Industrial Revolution (4IR or Industry 40), the digital world has a wealth of data, such as Internet of Things (IoT) data, cybersecurity data, mobile data, business data, social media data, health data, etc To intelligently analyze these data and develop the corresponding smart and automated applications, the knowledge of artificial intelligence (AI), particularly, machine learning (ML) is the key Various types of machine learning algorithms such as supervised, unsupervised, semi-supervised, and reinforcement learning exist in the area Besides, the deep learning, which is part of a broader family of machine learning methods, can intelligently analyze the data on a large scale In this paper, we present a comprehensive view on these machine learning algorithms that can be applied to enhance the intelligence and the capabilities of an application Thus, this study’s key contribution is explaining the principles of different machine learning techniques and their applicability in various real-world application domains, such as cybersecurity systems, smart cities, healthcare, e-commerce, agriculture, and many more We also highlight the challenges and potential research directions based on our study Overall, this paper aims to serve as a reference point for both academia and industry professionals as well as for decision-makers in various real-world situations and application areas, particularly from the technical point of view

Abstract: We introduce a general-purpose conditioning method for neural networks called FiLM: Feature-wise Linear Modulation. FiLM layers influence neural network computation via a simple, feature-wise affine transformation based on conditioning information. We show that FiLM layers are highly effective for visual reasoning - answering image-related questions which require a multi-step, high-level process - a task which has proven difficult for standard deep learning methods that do not explicitly model reasoning. Specifically, we show on visual reasoning tasks that FiLM layers 1) halve state-of-the-art error for the CLEVR benchmark, 2) modulate features in a coherent manner, 3) are robust to ablations and architectural modifications, and 4) generalize well to challenging, new data from few examples or even zero-shot.

Abstract: In this paper, we address the task of natural language object retrieval, to localize a target object within a given image based on a natural language query of the object. Natural language object retrieval differs from text-based image retrieval task as it involves spatial information about objects within the scene and global scene context. To address this issue, we propose a novel Spatial Context Recurrent ConvNet (SCRC) model as scoring function on candidate boxes for object retrieval, integrating spatial configurations and global scene-level contextual information into the network. Our model processes query text, local image descriptors, spatial configurations and global context features through a recurrent network, outputs the probability of the query text conditioned on each candidate box as a score for the box, and can transfer visual-linguistic knowledge from image captioning domain to our task. Experimental results demonstrate that our method effectively utilizes both local and global information, outperforming previous baseline methods significantly on different datasets and scenarios, and can exploit large scale vision and language datasets for knowledge transfer.

Abstract: Suppression effect in multiple regression analysis may be more common in research than what is currently recognized. We have reviewed several literatures of interest which treats the concept and types of suppressor variables. Also, we have highlighted systematic ways to identify suppression effect in multiple regressions using statistics such as: R2, sum of squares, regression weight and comparing zero-order correlations with Variance Inflation Factor (VIF) respectively. We also establish that suppression effect is a function of multicollinearity; however, a suppressor variable should only be allowed in a regression analysis if its VIF is less than five (5).

Abstract: Finding the fastest pathway to vaccine availability includes the avoidance of safety pitfalls

Abstract: Deep artificial neural networks (DNNs) are typically trained via gradient-based learning algorithms, namely backpropagation. Evolution strategies (ES) can rival backprop-based algorithms such as Q-learning and policy gradients on challenging deep reinforcement learning (RL) problems. However, ES can be considered a gradient-based algorithm because it performs stochastic gradient descent via an operation similar to a finite-difference approximation of the gradient. That raises the question of whether non-gradient-based evolutionary algorithms can work at DNN scales. Here we demonstrate they can: we evolve the weights of a DNN with a simple, gradient-free, population-based genetic algorithm (GA) and it performs well on hard deep RL problems, including Atari and humanoid locomotion. The Deep GA successfully evolves networks with over four million free parameters, the largest neural networks ever evolved with a traditional evolutionary algorithm. These results (1) expand our sense of the scale at which GAs can operate, (2) suggest intriguingly that in some cases following the gradient is not the best choice for optimizing performance, and (3) make immediately available the multitude of neuroevolution techniques that improve performance. We demonstrate the latter by showing that combining DNNs with novelty search, which encourages exploration on tasks with deceptive or sparse reward functions, can solve a high-dimensional problem on which reward-maximizing algorithms (e.g.\ DQN, A3C, ES, and the GA) fail. Additionally, the Deep GA is faster than ES, A3C, and DQN (it can train Atari in ${\raise.17ex\hbox{$\scriptstyle\sim$}}$4 hours on one desktop or ${\raise.17ex\hbox{$\scriptstyle\sim$}}$1 hour distributed on 720 cores), and enables a state-of-the-art, up to 10,000-fold compact encoding technique.

Abstract: PURPOSEIn KEYNOTE-189, first-line pembrolizumab plus pemetrexed-platinum significantly improved overall survival (OS) and progression-free survival (PFS) compared with placebo plus pemetrexed-plati...

Abstract: It was only a matter of time before deep neural networks (DNNs) – deep learning – made their mark in turbulence modelling, or more broadly, in the general area of high-dimensional, complex dynamical systems. In the last decade, DNNs have become a dominant data mining tool for big data applications. Although neural networks have been applied previously to complex fluid flows, the article featured here (Ling et al., J. Fluid Mech., vol. 807, 2016, pp. 155–166) is the first to apply a true DNN architecture, specifically to Reynolds averaged Navier Stokes turbulence models. As one often expects with modern DNNs, performance gains are achieved over competing state-of-the-art methods, suggesting that DNNs may play a critically enabling role in the future of modelling complex flows.

Abstract: Correlation filter (CF) based trackers have recently gained a lot of popularity due to their impressive performance on benchmark datasets, while maintaining high frame rates. A significant amount of recent research focuses on the incorporation of stronger features for a richer representation of the tracking target. However, this only helps to discriminate the target from background within a small neighborhood. In this paper, we present a framework that allows the explicit incorporation of global context within CF trackers. We reformulate the original optimization problem and provide a closed form solution for single and multi-dimensional features in the primal and dual domain. Extensive experiments demonstrate that this framework significantly improves the performance of many CF trackers with only a modest impact on frame rate.