Thursday, 26 November 2020

02:02 AM

Adversarial Evaluation of Multimodal Models under Realistic Gray Box Assumption. (arXiv:2011.12902v1 [cs.CV]) [cs.CR updates on]

This work examines the vulnerability of multimodal (image + text) models to adversarial threats similar to those discussed in previous literature on unimodal (image- or text-only) models. We introduce realistic assumptions of partial model knowledge and access, and discuss how these assumptions differ from the standard "black-box"/"white-box" dichotomy common in current literature on adversarial attacks. Working under various levels of these "gray-box" assumptions, we develop new attack methodologies unique to multimodal classification and evaluate them on the Hateful Memes Challenge classification task. We find that attacking multiple modalities yields stronger attacks than unimodal attacks alone (inducing errors in up to 73% of cases), and that the unimodal image attacks on multimodal classifiers we explored were stronger than character-based text augmentation attacks (inducing errors on average in 45% and 30% of cases, respectively).

EI-MTD:Moving Target Defense for Edge Intelligence against Adversarial Attacks. (arXiv:2009.10537v3 [cs.CR] UPDATED) [cs.CR updates on]

With the boom of edge intelligence, its vulnerability to adversarial attacks becomes an urgent problem. The so-called adversarial example can fool a deep learning model on the edge node to misclassify. Due to the property of transferability, the adversary can easily make a black-box attack using a local substitute model. Nevertheless, the limitation of resource of edge nodes cannot afford a complicated defense mechanism as doing on the cloud data center. To overcome the challenge, we propose a dynamic defense mechanism, namely EI-MTD. It first obtains robust member models with small size through differential knowledge distillation from a complicated teacher model on the cloud data center. Then, a dynamic scheduling policy based on a Bayesian Stackelberg game is applied to the choice of a target model for service. This dynamic defense can prohibit the adversary from selecting an optimal substitute model for black-box attacks. Our experimental result shows that this dynamic scheduling can effectively protect edge intelligence against adversarial attacks under the black-box setting.

Wednesday, 25 November 2020

06:02 PM

Burning questions about smouldering myeloma [Nature - Issue - science feeds]

Nature, Published online: 25 November 2020; doi:10.1038/d41586-020-03225-0

Researchers are amassing evidence about the best ways to treat the precursor condition before it develops into active disease.

Progenitor identification and SARS-CoV-2 infection in human distal lung organoids [Nature - Issue - science feeds]

Nature, Published online: 25 November 2020; doi:10.1038/s41586-020-3014-1

Progenitor identification and SARS-CoV-2 infection in human distal lung organoids

Scientists are harnessing viruses to treat tumours [Nature - Issue - science feeds]

Nature, Published online: 25 November 2020; doi:10.1038/d41586-020-03226-z

Could viruses such as the one used in the measles vaccine be used to treat multiple myeloma?

Fibrosis: from mechanisms to medicines [Nature - Issue - science feeds]

Nature, Published online: 25 November 2020; doi:10.1038/s41586-020-2938-9

This review discusses how single-cell profiling and other technological advances are increasing our understanding of the mechanisms of fibrosis, thereby accelerating the discovery, development and testing of new treatments.

Are we the same person throughout our lives? In essence, yes [EurekAlert! - Breaking News]

Although our body changes and our beliefs and values may vary throughout our lives, our essence remains stable. Research at the Complutense University of Madrid (UCM) has recorded brain activity in a group of individuals showing that our ability to recognise ourselves as distinctive --the "continuity of the self"-- remains undiminished by change and that it takes us only 250 milliseconds to recognise ourselves.

New insights into how the CRISPR immune system evolved [EurekAlert! - Breaking News]

With new insights into how the genetic tool CRISPR - which allows direct editing of our genes - evolved and adapted, we are now one step closer to understanding the basis of the constant struggle for survival that takes place in nature. The results can be used in future biotechnologies.

Iron infusion proves effective to treat anaemia in Rural Africa [EurekAlert! - Breaking News]

Iron-deficiency anaemia is a major concern in low-income settings, especially for women. In a new study by the Swiss Tropical and Public Health Institute (Swiss TPH) and partners published yesterday in The Lancet Global Health, researchers found that iron infusion was feasible, safe and, in contrast to the standard iron-deficiency anaemia treatment of oral iron tablets, highly effective in Tanzania. This is the first study to provide evidence of the benefits and safety of iron infusion in a low-income setting.

Grabbing viruses out of thin air [EurekAlert! - Breaking News]

Materials that convert mechanical into electrical or magnetic energy could open the door to a future of wearable and structure-integrated virus sensors.

Stem cell-based screen identifies potential new treatments [EurekAlert! - Breaking News]

In a recent study published in Stem Cell Reports, Seba Almedawar, PhD, and colleagues with the Center for Regenerative Therapies TU Dresden, Germany, used induced pluripotent stem cells (iPSCs) derived from the skin of healthy donors and of patients with retinitis pigmentosa to find drugs with the potential to enhance RPE phagocytosis.

02:02 AM

Nudge Attacks on Point-Cloud DNNs. (arXiv:2011.11637v1 [cs.CR]) [cs.CR updates on]

The wide adaption of 3D point-cloud data in safety-critical applications such as autonomous driving makes adversarial samples a real threat. Existing adversarial attacks on point clouds achieve high success rates but modify a large number of points, which is usually difficult to do in real-life scenarios. In this paper, we explore a family of attacks that only perturb a few points of an input point cloud, and name them nudge attacks. We demonstrate that nudge attacks can successfully flip the results of modern point-cloud DNNs. We present two variants, gradient-based and decision-based, showing their effectiveness in white-box and grey-box scenarios. Our extensive experiments show nudge attacks are effective at generating both targeted and untargeted adversarial point clouds, by changing a few points or even a single point from the entire point-cloud input. We find that with a single point we can reliably thwart predictions in 12--80% of cases, whereas 10 points allow us to further increase this to 37--95%. Finally, we discuss the possible defenses against such attacks, and explore their limitations.

Differentially Private Learning Needs Better Features (or Much More Data). (arXiv:2011.11660v1 [cs.LG]) [cs.CR updates on]

We demonstrate that differentially private machine learning has not yet reached its "AlexNet moment" on many canonical vision tasks: linear models trained on handcrafted features significantly outperform end-to-end deep neural networks for moderate privacy budgets. To exceed the performance of handcrafted features, we show that private learning requires either much more private data, or access to features learned on public data from a similar domain. Our work introduces simple yet strong baselines for differentially private learning that can inform the evaluation of future progress in this area.

On The Round Complexity of Two-Party Quantum Computation. (arXiv:2011.11212v1 [quant-ph] CROSS LISTED) [cs.CR updates on]

We investigate the round complexity of maliciously-secure two-party quantum computation (2PQC) with setup, and obtain the following results:

- A three-message protocol (two-message if only one party receives output) in the common random string (CRS) model assuming classical two-message oblivious transfer (OT) with post-quantum malicious security. This round complexity is optimal for the sequential communication setting. Under the additional assumption of reusable malicious designated-verifier non-interactive zero-knowledge (MDV-NIZK) arguments for NP, our techniques give an MDV-NIZK for QMA. Each of the assumptions mentioned above is known from the quantum hardness of learning with errors (QLWE).

- A protocol with two simultaneous rounds of communication, in a quantum preprocessing model, assuming sub-exponential QLWE. In fact, we construct a three-round protocol in the CRS model with only two rounds of online communication, which implies the above result. Along the way, we develop a new delayed simulation technique that we call "simulation via teleportation," which may be useful in other settings.

In addition, we perform a preliminary investigation into barriers and possible approaches for two-round 2PQC in the CRS model, including an impossibility result for a natural class of simulators, and a proof-of-concept construction from a strong form of quantum virtual black-box (VBB) obfuscation.

Prior to our work, maliciously-secure 2PQC required round complexity linear in the size of the quantum circuit.

Tuesday, 24 November 2020

06:02 PM

Active Surveillance Testing to Prevent Staph Spread in NICUs [JAMA Current Issue]

Personnel in neonatal intensive care units (NICUs) should conduct active surveillance testing for Staphylococcus aureus colonization of neonates to prevent the spread of infection during outbreaks or when incidence increases, according to a new CDC guideline.

Opportunistic Atrial Fibrillation Screening Doesn’t Catch More Cases [JAMA Current Issue]

Dutch primary care practices that systematically tested older adults for atrial fibrillation (AF) did not detect more AF than usual care, according to a trial reported in the BMJ.

Long-term Mortality Outcomes in Early Breast Cancer Screening Trial [JAMA Current Issue]

Final results from the 1990s UK Age trial suggest that yearly mammography starting around age 40 years could potentially reduce breast cancer mortality with minimal overdiagnosis. However, research is needed to clarify whether current screening technologies and treatments would reduce the mortality benefit observed with earlier screening, the authors stated in Lancet Oncology.

Clinical Genome Sequencing—The Importance of Implementation Data [JAMA Current Issue]

This Viewpoint discusses the importance of understanding how next-generation genomic sequencing is being implemented—for whom noninvasive prenatal testing, whole-exome and -genome sequencing for suspected genetic disorders, and tumor sequencing is being used and who is paying for it—as precondition for knowing how the technologies can best serve patients.

Herd Immunity and Implications for SARS-CoV-2 Control [JAMA Current Issue]

This JAMA Insights Clinical Update discusses herd immunity in the context of the COVID-19 pandemic and explains the herd immunity threshold as a function of transmissibility (R0), the role of an effective vaccine and vaccination program in sustaining herd immunity, and the risks of an infection-based herd immunity approach.

10:03 AM

A structurally conserved human and Tetrahymena telomerase catalytic core [Biophysics and Computational Biology] [Early Edition]

Telomerase is a ribonucleoprotein complex that counteracts the shortening of chromosome ends due to incomplete replication. Telomerase contains a catalytic core of telomerase reverse transcriptase (TERT) and telomerase RNA (TER). However, what defines TERT and separates it from other reverse transcriptases remains a subject of debate. A recent cryoelectron microscopy...

02:03 AM

Daily briefing: Oxford–AstraZeneca COVID vaccine works — but scientists have questions [Nature - Issue - science feeds]

Nature, Published online: 23 November 2020; doi:10.1038/d41586-020-03328-8

Early data indicate that the Oxford–AstraZeneca jab is effective, but dose makes a difference. Plus, the scientific dilemma posed by emergency vaccine approvals, and an AI that sums up papers in a sentence.

iCmSC: Incomplete Cross-Modal Subspace Clustering [IEEE Transactions on Image Processing - new TOC]

Cross-modal clustering aims to cluster the high-similar cross-modal data into one group while separating the dissimilar data. Despite the promising cross-modal methods have developed in recent years, existing state-of-the-arts cannot effectively capture the correlations between cross-modal data when encountering with incomplete cross-modal data, which can gravely degrade the clustering performance. To well tackle the above scenario, we propose a novel incomplete cross-modal clustering method that integrates canonical correlation analysis and exclusive representation, named incomplete Cross-modal Subspace Clustering (i.e., iCmSC). To learn a consistent subspace representation among incomplete cross-modal data, we maximize the intrinsic correlations among different modalities by deep canonical correlation analysis (DCCA), while an exclusive self-expression layer is proposed after the output layers of DCCA. We exploit a $ell _{1,2}$ -norm regularization in the learned subspace to make the learned representation more discriminative, which makes samples between different clusters mutually exclusive and samples among the same cluster attractive to each other. Meanwhile, the decoding networks are employed to reconstruct the feature representation, and further preserve the structural information among the original cross-modal data. To the end, we demonstrate the effectiveness of the proposed iCmSC via extensive experiments, which can justify that iCmSC achieves consistently large improvement compared with the state-of-the-arts.

Friday, 20 November 2020

10:02 AM

Machine-learning iterative calculation of entropy for physical systems [Physics] [Early Edition]

Characterizing the entropy of a system is a crucial, and often computationally costly, step in understanding its thermodynamics. It plays a key role in the study of phase transitions, pattern formation, protein folding, and more. Current methods for entropy estimation suffer from a high computational cost, lack of generality, or...

02:03 AM

[Comment] Never too old to benefit from lipid-lowering treatment [The Lancet]

In The Lancet, Martin Bødtker Mortensen and Børge Grønne Nordestgaard1 report the results of a large cohort of over 90 000 individuals aged 20–100 years, from the Copenhagen General Population Study, who were not on statin therapy and did not have atherosclerotic cardiovascular disease or diabetes. Within this primary prevention cohort, higher LDL-cholesterol concentrations were associated with an increased absolute risk of myocardial infarction and atherosclerotic cardiovascular disease in those aged 70 years or older and the number needed to treat to prevent one myocardial infarction in 5 years was substantially lower in the individuals aged 70–100 years compared with that in younger individuals.

[Comment] Offline: Holocaust education—a medical imperative [The Lancet]

As school students in England in the 1970s, we were taught the bare facts of the Holocaust. And that is where we left it. A fact of history, a fact certainly to be remembered, yet a fact that seemed very distant from contemporary times. But last week, the Ontario Medical Association, together with Doctors Against Racism and Anti-Semitism, held a webinar to discuss the case for teaching the Holocaust in medical schools. The meeting was led by Dr Frank Sommers, a professor of psychiatry at the University of Toronto.

[World Report] Violence leads to health emergencies in Mozambique [The Lancet]

Displaced by war, thousands of people in Mozambique are at risk of infectious disease. Munyaradzi Makoni reports.

[Perspectives] The banality of the patriarchy [The Lancet]

When German–American philosopher Hannah Arendt covered the 1961 trial of the Nazi Adolf Eichmann for The New Yorker magazine—her reports led to her book of 1963, Eichmann in Jerusalem: A Report on the Banality of Evil—she was struck by the concept of banality. Arendt wrote that Eichmann, this bureaucrat who was “medium-sized, slender, middle-aged, with receding hair, ill-fitting teeth, and nearsighted eyes” was responsible for participation in one of the greatest crimes in history. The banality to which Arendt referred, however, was not Eichmann's unexceptional physical appearance, but “an inability to think; that is, to think from the standpoint of somebody else”.

[Correspondence] Institutional versus home isolation to curb the COVID-19 outbreak – Authors' reply [The Lancet]

We thank Ajeet Singh Bhadoria and colleagues for their insightful comments in response to our Correspondence.1 Although we focus on teasing out the impact of institutional isolation beyond the other non-pharmaceutical measures, we agree that isolation of all cases cannot be a stand-alone strategy. The pandemic response to COVID-19 must be a multipronged approach that includes liberal testing, tracing and quarantine of contacts, physical distancing, and widespread use of face masks—such a multipronged approach is particularly crucial for a disease with a high asymptomatic rate.

Hierarchical and Interactive Refinement Network for Edge-Preserving Salient Object Detection [IEEE Transactions on Image Processing - new TOC]

Salient object detection has undergone a very rapid development with the blooming of Deep Neural Network (DNN), which is usually taken as an important preprocessing procedure in various computer vision tasks. However, the down-sampling operations, such as pooling and striding, always make the final predictions blurred at edges, which has seriously degenerated the performance of salient object detection. In this paper, we propose a simple yet effective approach, i.e., Hierarchical and Interactive Refinement Network (HIRN), to preserve the edge structures in detecting salient objects. In particular, a novel multi-stage and dual-path network structure is designed to estimate the salient edges and regions from the low-level and high-level feature maps, respectively. As a result, the predicted regions will become more accurate by enhancing the weak responses at edges, while the predicted edges will become more semantic by suppressing the false positives in background. Once the salient maps of edges and regions are obtained at the output layers, a novel edge-guided inference algorithm is introduced to further filter the resulting regions along the predicted edges. Extensive experiments on several benchmark datasets have been conducted, in which the results show that our method significantly outperforms a variety of state-of-the-art approaches.

Learnable Descriptors for Visual Search [IEEE Transactions on Image Processing - new TOC]

This work proposes LDVS, a learnable binary local descriptor devised for matching natural images within the MPEG CDVS framework. LDVS descriptors are learned so that they can be sign-quantized and compared using the Hamming distance. The underlying convolutional architecture enjoys a moderate parameters count for operations on mobile devices. Our experiments show that LDVS descriptors perform favorably over comparable learned binary descriptors at patch matching on two different datasets. A complete pair-wise image matching pipeline is then designed around LDVS descriptors, integrating them in the reference CDVS evaluation framework. Experiments show that LDVS descriptors outperform the compressed CDVS SIFT-like descriptors at pair-wise image matching over the challenging CDVS image dataset.

Fast Roughness Minimizing Image Restoration Under Mixed Poisson–Gaussian Noise [IEEE Transactions on Image Processing - new TOC]

Image acquisition in many biomedical imaging modalities is corrupted by Poisson noise followed by additive Gaussian noise. While total variation and related regularization methods for solving biomedical inverse problems are known to yield high quality reconstructions in most situations, such methods mostly use log-likelihood of either Gaussian or Poisson noise models, and rarely use mixed Poisson-Gaussian (PG) noise model. There is a recent work which deals with exact PG likelihood and total variation regularization. This method adapts the primal-dual approach involving gradients steps on the PG log-likelihood, with step size limited by the inverse of its Lipschitz constant. This leads to limitations in the convergence speed. Although the alternating direction method of multipliers (ADMM) algorithm does not have such step size restrictions, ADMM has never been applied for this problem, for the possible reason that PG log-likelihood is quite complex. In this paper, we develop an ADMM based optimization for roughness minimizing image restoration under PG log-likelihood. We achieve this by first developing a novel iterative method for computing the proximal solution of PG log-likelihood, deriving the termination conditions for this iterative method, and then integrating into a provably convergent ADMM scheme. We experimentally demonstrate that the proposed method outperform primal-dual method in most of the cases.

Learning Dual Encoding Model for Adaptive Visual Understanding in Visual Dialogue [IEEE Transactions on Image Processing - new TOC]

Different from Visual Question Answering task that requires to answer only one question about an image, Visual Dialogue task involves multiple rounds of dialogues which cover a broad range of visual content that could be related to any objects, relationships or high-level semantics. Thus one of the key challenges in Visual Dialogue task is to learn a more comprehensive and semantic-rich image representation that can adaptively attend to the visual content referred by variant questions. In this paper, we first propose a novel scheme to depict an image from both visual and semantic views. Specifically, the visual view aims to capture the appearance-level information in an image, including objects and their visual relationships, while the semantic view enables the agent to understand high-level visual semantics from the whole image to the local regions. Furthermore, on top of such dual-view image representations, we propose a Dual Encoding Visual Dialogue (DualVD) module, which is able to adaptively select question-relevant information from the visual and semantic views in a hierarchical mode. To demonstrate the effectiveness of DualVD, we propose two novel visual dialogue models by applying it to the Late Fusion framework and Memory Network framework. The proposed models achieve state-of-the-art results on three benchmark datasets. A critical advantage of the DualVD module lies in its interpretability. We can analyze which modality (visual or semantic) has more contribution in answering the current question by explicitly visualizing the gate values. It gives us insights in understanding of information selection mode in the Visual Dialogue task. The code is available at

An Overview of Recent Advances in Coordinated Control of Multiple Autonomous Surface Vehicles [IEEE Transactions on Industrial Informatics - new TOC]

Autonomous surface vehicles (ASVs) are marine vessels capable of performing various marine operations without a crew in a variety of cluttered and hostile water/ocean environments. For complex missions, there are increasing needs for deploying a fleet of ASVs instead of a single one to complete difficult tasks. Cooperative operations with a fleet of ASVs offer great advantages with enhanced capability and efficacy. Despite various application potentials, coordinated motion control of ASVs pose great challenges due to the multiplicity of ASVs, complexity of intravehicle interactions and fleet formation with collision avoidance requirements, and scarcity of communication bandwidths in sea environments. Coordinated control of multiple ASVs has received considerable attention in the last decade. This article provides an overview of recent advances in coordinated control of multiple ASVs. First, some challenging issues and scenarios in motion control of ASVs are presented. Next, coordinated control architecture and methods of multiple ASVs are briefly discussed. Then, recent results on trajectory-guided, path-guided, and target-guided coordinated control of multiple ASVs are reviewed in detail. Finally, several theoretical and technical issues are suggested to direct future investigations including network-based coordination, event-triggered coordination, collision-free coordination, optimization-based coordination, data-driven coordination of ASVs, and task-region-oriented coordination of multiple ASVs and autonomous underwater vehicles.

Reinforcement Learning for Real-Time Pricing and Scheduling Control in EV Charging Stations [IEEE Transactions on Industrial Informatics - new TOC]

This article proposes a reinforcement-learning (RL) approach for optimizing charging scheduling and pricing strategies that maximize the system objective of a public electric vehicle (EV) charging station. The proposed algorithm is “online” in the sense that the charging and pricing decisions made at each time depend only on the observation of past events, and is “model-free” in the sense that the algorithm does not rely on any assumed stochastic models of uncertain events. To cope with the challenge arising from the time-varying continuous state and action spaces in the RL problem, we first show that it suffices to optimize the total charging rates to fulfill the charging requests before departure times. Then, we propose a feature-based linear function approximator for the state–value function to further enhance the efficiency and generalization ability of the proposed algorithm. Through numerical simulations with real-world data, we show that the proposed RL algorithm achieves on average 138.5% higher charging-station profit than representative benchmark algorithms.

Remaining Useful Life Prediction Using a Novel Feature-Attention-Based End-to-End Approach [IEEE Transactions on Industrial Informatics - new TOC]

Deep learning plays an increasingly important role in industrial applications, such as the remaining useful life (RUL) prediction of machines. However, when dealing with multifeature data, most deep learning approaches do not have effective mechanisms to weigh the input features adaptively. In this article, a novel feature-attention-based end-to-end approach is proposed for RUL prediction. First, the proposed feature-attention mechanism is directly applied to the input data, which gives greater attention weights to more important features dynamically in the training process. This helps the model focus more on those critical inputs, and the prediction performance is therefore improved. Next, bidirectional gated recurrent units (BGRU) are used to extract long-term dependencies from the weighted input data, and convolutional neural networks are employed to capture local features from the output sequences of BGRU. Finally, fully connected networks are used to learn the above-mentioned abstract representations to predict the RUL. The proposed approach is validated in a case study of turbofan engines. The experimental results demonstrate that the proposed approach outperforms other latest existing approaches.

Adaptive Multivariate Data Compression in Smart Metering Internet of Things [IEEE Transactions on Industrial Informatics - new TOC]

Recent advances in electric metering infrastructure have given rise to the generation of gigantic chunks of data. Transmission of all of these data certainly poses a significant challenge in bandwidth and storage constrained Internet of Things (IoT), where smart meters act as sensors. In this work, a novel multivariate data compression scheme is proposed for smart metering IoT. The proposed algorithm exploits the cross correlation between different variables sensed by smart meters to reduce the dimension of data. Subsequently, sparsity in each of the decorrelated streams is utilized for temporal compression. To examine the quality of compression, the multivariate data is characterized using multivariate normal–autoregressive integrated moving average modeling before compression as well as after reconstruction of the compressed data. Our performance studies indicate that compared to the state-of-the-art, the proposed technique is able to achieve impressive bandwidth saving for transmission of data over communication network without compromising faithful reconstruction of data at the receiver. The proposed algorithm is tested in a real smart metering setup and its time complexity is also analyzed.

Vehicle to Grid Frequency Regulation Capacity Optimal Scheduling for Battery Swapping Station Using Deep Q-Network [IEEE Transactions on Industrial Informatics - new TOC]

Battery swapping stations (BSSs) are ideal candidates for fast frequency regulation services (FFRS) due to their large battery stock capacity. In addition, BSSs can precharge batteries for customers and the batteries that are not in charging can provide a stable regulation capacity to the market. However, uncertainties, such as ACE signals and the EV per-hour visit counts, introduce stochastic nonlinear dynamics into the operation of a BSS-based FFRS. Currently, there is no quantification method to ensure its optimal economical operation. To close this gap, in this article, we propose a novel deep Q-learning-based FFRS capacity dynamic scheduling strategy. This method can autonomously schedule the hourly regulation capacity in real time to maximize the BSS's revenue for providing FFRS. Case studies using real-world data verify the efficacy of the proposed work.

Thursday, 19 November 2020

10:03 AM

Single-molecule and in silico dissection of the interaction between Polycomb repressive complex 2 and chromatin [Biophysics and Computational Biology] [Early Edition]

Polycomb repressive complex 2 (PRC2) installs and spreads repressive histone methylation marks on eukaryotic chromosomes. Because of the key roles that PRC2 plays in development and disease, how this epigenetic machinery interacts with DNA and nucleosomes is of major interest. Nonetheless, the mechanism by which PRC2 engages with native-like chromatin...

Tuesday, 17 November 2020

02:02 AM

Thirst recruits phasic dopamine signaling through subfornical organ neurons [Neuroscience] [Early Edition]

Thirst is a highly potent drive that motivates organisms to seek out and consume balance-restoring stimuli. The detection of dehydration is well understood and involves signals of peripheral origin and the sampling of internal milieu by first order homeostatic neurons within the lamina terminalis—particularly glutamatergic neurons of the subfornical organ...

Correction for Kessler et al., Victorin, the host-selective cyclic peptide toxin from the oat pathogen Cochliobolus victoriae, is ribosomally encoded [Corrections] [Early Edition]

BIOCHEMISTRY Correction for “Victorin, the host-selective cyclic peptide toxin from the oat pathogen Cochliobolus victoriae, is ribosomally encoded,” by Simon C. Kessler, Xianghui Zhang, Megan C. McDonald, Cameron L. M. Gilchrist, Zeran Lin, Adriana Rightmyer, Peter S. Solomon, B. Gillian Turgeon, and Yit-Heng Chooi, which was first published September 14,...

When Automatic Voice Disguise Meets Automatic Speaker Verification [IEEE Transactions on Information Forensics and Security - new TOC]

The technique of transforming voices in order to hide the real identity of a speaker is called voice disguise, among which automatic voice disguise (AVD) by modifying the spectral and temporal characteristics of voices with miscellaneous algorithms are easily conducted with softwares accessible to the public. AVD has posed great threat to both human listening and automatic speaker verification (ASV). In this paper, we have found that ASV is not only a victim of AVD but could be a tool to beat some simple types of AVD. Firstly, three types of AVD, pitch scaling, vocal tract length normalization (VTLN) and voice conversion (VC), are introduced as representative methods. State-of-the-art ASV methods are subsequently utilized to objectively evaluate the impact of AVD on ASV by equal error rates (EER). Moreover, an approach to restore disguised voice to its original version is proposed by minimizing a function of ASV scores w.r.t. restoration parameters. Experiments are then conducted on disguised voices from Voxceleb, a dataset recorded in real-world noisy scenario. The results have shown that, for the voice disguise by pitch scaling, the proposed approach obtains an EER around 7% comparing to the 30% EER of a recently proposed baseline using the ratio of fundamental frequencies. The proposed approach generalizes well to restore the disguise with nonlinear frequency warping in VTLN by reducing its EER from 34.3% to 18.5%. However, it is difficult to restore the source speakers in VC by our approach, where more complex forms of restoration functions or other paralinguistic cues might be necessary to restore the nonlinear transform in VC. Finally, contrastive visualization on ASV features with and without restoration illustrate the role of the proposed approach in an intuitive way.

High Intrinsic Dimensionality Facilitates Adversarial Attack: Theoretical Evidence [IEEE Transactions on Information Forensics and Security - new TOC]

Machine learning systems are vulnerable to adversarial attack. By applying to the input object a small, carefully-designed perturbation, a classifier can be tricked into making an incorrect prediction. This phenomenon has drawn wide interest, with many attempts made to explain it. However, a complete understanding is yet to emerge. In this paper we adopt a slightly different perspective, still relevant to classification. We consider retrieval, where the output is a set of objects most similar to a user-supplied query object, corresponding to the set of k-nearest neighbors. We investigate the effect of adversarial perturbation on the ranking of objects with respect to a query. Through theoretical analysis, supported by experiments, we demonstrate that as the intrinsic dimensionality of the data domain rises, the amount of perturbation required to subvert neighborhood rankings diminishes, and the vulnerability to adversarial attack rises. We examine two modes of perturbation of the query: either `closer' to the target point, or `farther' from it. We also consider two perspectives: `query-centric', examining the effect of perturbation on the query's own neighborhood ranking, and `target-centric', considering the ranking of the query point in the target's neighborhood set. All four cases correspond to practical scenarios involving classification and retrieval.

NeuroAED: Towards Efficient Abnormal Event Detection in Visual Surveillance With Neuromorphic Vision Sensor [IEEE Transactions on Information Forensics and Security - new TOC]

Abnormal event detection is an important task in research and industrial applications, which has received considerable attention in recent years. Existing methods usually rely on standard frame-based cameras to record the data and process them with computer vision technologies. In contrast, this paper presents a novel neuromorphic vision based abnormal event detection system. Compared to the frame-based camera, neuromorphic vision sensors, such as Dynamic Vision Sensor (DVS), do not acquire full images at a fixed frame rate but rather have independent pixels that output intensity changes (called events) asynchronously at the time they occur. Thus, it avoids the design of the encryption scheme. Since events are triggered by moving edges on the scene, DVS is a natural motion detector for the abnormal objects and automatically filters out any temporally-redundant information. Based on this unique output, we first propose a highly efficient method based on the event density to select activated event cuboids and locate the foreground. We design a novel event-based multiscale spatio-temporal descriptor to extract features from the activated event cuboids for the abnormal event detection. Additionally, we build the NeuroAED dataset, the first public dataset dedicated to abnormal event detection with neuromorphic vision sensor. The NeuroAED dataset consists of four sub-datasets: Walking, Campus, Square, and Stair dataset. Experiments are conducted based on these datasets and demonstrate the high efficiency and accuracy of our method.

Comments on “Privacy-Preserving Public Auditing Protocol for Regenerating-Code-Based Cloud Storage” [IEEE Transactions on Information Forensics and Security - new TOC]

Public auditing protocol is crucial for the success of cloud computing, as it can ensure the outsourced data in cloud server are not tampered by attackers. Due to its importance, public auditing protocol has received considerable attention in the past years. In 2015, Liu et al. proposed a privacy-preserving public auditing protocol for regenerating-code-based cloud storage (IEEE Transactions on Information Forensics and Security, 10(7):1513–1528, 2015) and claimed it is secure under the considered security model. However, in this article, we will show that their protocol is not as secure as they claimed, i.e., the proxy delegated by the data owner can forge an authenticator for any data block, which obviously invalidates their protocol’s security. We hope that by identifying the design flaw, similar weaknesses can be avoided in future protocol design.

Thursday, 12 November 2020

Wednesday, 11 November 2020

10:02 AM

Career Performance Trajectories of Professional Australian Football Players [International Journal of Sports Physiology and Performance]

Journal Name: International Journal of Sports Physiology and Performance
Volume: 15
Issue: 10
Pages: 1363-1368

Workload Differences Between Training Drills and Competition in Elite Netball [International Journal of Sports Physiology and Performance]

Journal Name: International Journal of Sports Physiology and Performance
Volume: 15
Issue: 10
Pages: 1385-1392

Response of Blood Biomarkers to Sprint Interval Swimming [International Journal of Sports Physiology and Performance]

Journal Name: International Journal of Sports Physiology and Performance
Volume: 15
Issue: 10
Pages: 1442-1447

Effects of Time of Day on Pacing in a 4-km Time Trial in Trained Cyclists [International Journal of Sports Physiology and Performance]

Journal Name: International Journal of Sports Physiology and Performance
Volume: 15
Issue: 10
Pages: 1455-1459

Friday, 06 November 2020

02:03 AM

Optimal Design of a Dual Stator Winding Induction Motor With Minimum Rate Reduction Level [IEEE Transactions on Industrial Electronics - new TOC]

Flux estimation of single winding induction machines (SWIMs) operating in a zero-speed region is challenging, since in this operation region, the voltage drop on the stator resistance value, which varies with the operating point, is not negligible. Furthermore, in this operation region, because of the interference between higher band frequencies of a dc-rejection filter with low-pass filter of the stator currents, the dc-offset of measured currents may distort the true values of the stator current. Dual stator winding induction machines (DSWIMs), which have two sets of three-phase windings with different number of pole pairs and adapt a standard squirrel cage rotor, have overcame the associated problems of SWIMs in the zero-speed region. Nevertheless, DWSIMs have lower power rating than SWIMs if they use the same stator and rotor frames. In this article, an optimal design procedure is proposed for DSWIMs, which benefits the advantages of DSWIMs and surmounts the rate reduction problem as much as possible. In this regard, the optimal flux levels ratio of winding sets and the best pole pair ratio are determined first. Afterward, the winding specifications are designated to have maximum output power, while the DSWIM advantages, especially the zero-speed region operation capability, remain valid. A DSWIM is designed, simulated in ANSYS/MAXWELL and fabricated based on the proposed method and utilizing commercially available standard stator and rotor frames. Experimental assessments verify that the DSWIM prototype has improved the rate-reduction problem up to 19% in comparison to similar proposed DSWIMs.

Temperature Field Calculation and Water-Cooling Structure Design of Coreless Permanent Magnet Synchronous Linear Motor [IEEE Transactions on Industrial Electronics - new TOC]

In this article, a water-cooled structure with water-cooled plates between the primary double-layer windings field of the primary structure under short-term high overload conditions and periodic operations is calculated. Second, the heat dissipation effects of two different waterway topologies are designed and compared. Based on the thermal network analysis model and multiphysical field coupling model, the heat dissipation effect of the water-cooled motor is obtained. In order to obtain the maximum sustainable working current that the motor can withstand under water-cooled conditions, a fast calculation model—response surface model—is established and the electromagnetic performance of the water-cooled motor is obtained. The analysis results show that the temperature rise of the water-cooled motor decreases greatly, and the average thrust increases by 2.73 times. Finally, the calculated and tested results of the prototype are compared and analyzed.

A Model Predictive Control Method for Grid-Connected Power Converters Without AC Voltage Sensors [IEEE Transactions on Industrial Electronics - new TOC]

A model predictive control (MPC) strategy for grid-connected power converters is proposed in this article to reduce the current ripples without using ac voltage sensors. First, a sliding-mode observer is designed to estimate the grid voltage with an adaptive compensation algorithm. Thus, grid-voltage sensors are removed. The novelty of the proposed grid-voltage observation method is that it is inherent frequency adaptive without using the accurate grid angular frequency. By implementing the MPC strategy using the estimated grid voltage, the current ripples are reduced, especially under distorted grid-voltage conditions because of the low-pass filter used in the observer. Next, to further reduce the current ripples, a double-voltage vector-based MPC strategy is proposed based on the principle of the modulated MPC. Detailed theoretical analyses are also carried out to show its effectiveness for the first time. Finally, experimental studies are carried out to verify the validity of the proposed ac voltage sensorless MPC strategy.

Dynamic Neural Networks for Motion-Force Control of Redundant Manipulators: An Optimization Perspective [IEEE Transactions on Industrial Electronics - new TOC]

Accurate position-force control is a core and challenging problem in robotics, especially for manipulators with redundant degrees of freedom (DOFs). For example, trajectory tracking-based control usually fails for grinding robots due to intolerable impact forces imposed onto the end effectors. The main difficulties lie in the coupling of motion and contact force, redundancy resolution, physical constraints, etc. In this article, we propose a novel motion-force control strategy in the framework of projection recurrent neural networks (RNN). Tracking error and contact force are described in orthogonal spaces, respectively, and by selecting minimizing joint torque as secondary task, the control problem is formulated as a quadratic-programming (QP) problem under multiple constraints. In order to obtain real-time optimization of joint toque, which is nonconvex relative to joint angles, the original QP is reconstructed in the velocity level, where the original objective function is replaced by its time derivative. Then, a dynamic neural network, which is convergence provable is established to solve the modified QP problem online. This work generalizes projection RNN-based position control of manipulators to that of position-force control, which opens a new avenue to shift position-force control of manipulators from pure control perspective to cross design with both convergence and optimality consideration. Numerical and experimental results show that the proposed scheme achieves accurate position-force control, and is capable of handling inequality constraints such as joint angular, velocity, and torque limitations, simultaneously, consumption of joint torque can be decreased effectively.

Underwater Acoustic Sensor Networks With Cost Efficiency for Internet of Underwater Things [IEEE Transactions on Industrial Electronics - new TOC]

Despite the potential benefits of Internet of Underwater Things, a number of issues hinder its realization, including the need for communication reliability and cost-effectiveness. This article aims to optimize network design to implement cost-effective underwater acoustic sensor networks (UASNs) with 3D topology while supporting diverse communication quality of service (QoS) requirements. First, we present an analytical framework based on a queueing system that evaluates communication performances of UASNs, wherein each underwater sensor distributed within a 3D space under the sea surface performs fountain code (FC)-based automatic repeat request (ARQ) transmissions under the slotted-Aloha medium access control protocol. Under the proposed framework, we evaluate communication performances given in terms of successful FC-based ARQ transmission probability and the average queueing delay of an underwater sensor. When evaluating the performances, we formulate the service time of each underwater sensor as a function of network parameters, i.e., the density of data sink and amount of redundancy for FC-based ARQ transmission, before solving a function for accurate service time, such that each sensor can be represented by an M/G/1 queue. Further, our analysis can formulate an optimization problem that aims at minimizing total cost incurred to install and operate 3D UASNs, without compromising two communication QoS requirements. To solve this problem, we propose a recursive algorithm to approach an optimal solution in reasonable time. Numerical evaluations demonstrate the validity of the proposed algorithm.

Thursday, 29 October 2020

Friday, 23 October 2020

Tuesday, 13 October 2020

Friday, 09 October 2020

12:46 PM

Disarmed principals: institutional resilience and the non-enforcement of delegation [European Political Science Review]

Governments across the world increasingly rely on non-state agents for managing even the most sensitive tasks that range from running critical infrastructures to protecting citizens. While private agents frequently underperform, governments as principals tend nonetheless not to enforce delegation contracts. Why? We suggest the mechanism of institutional resilience. A preexisting set of rules shapes non-enforcement through the combination of (i) its structural misfit with the delegation contract and (ii) asymmetric interdependence that favors the agent over time. To demonstrate the plausibility of our argument, we trace the political process behind Europe’s largest military transport aircraft, the A400M. Governments delegated the development and production of this complex program to a private firm, Airbus. They layered a ‘commercial approach’ onto traditionally state-run defense industries. Yet, resilience caused these new formal rules to fail and eventually disarmed principals. Our mechanism constitutes an innovative approach by theorizing an alternative path toward dynamic continuity.

National identity, a blessing or a curse? The divergent links from national attachment, pride, and chauvinism to social and political trust [European Political Science Review]

Is it true that national identity increases trust, as liberal nationalists assume? Recent research has studied this side of the ‘national identity argument’ by focusing on conceptions of the content of national identity (often civic or ethnic) and their links to social, rather than political, trust. This paper argues that if we take social identity theory seriously, however, we need to complement this picture by asking how varying the strength – rather than the content – of a person’s sense of their national identity affects both their social and political trust. We break down the different dimensions of national identity, hypothesizing and empirically verifying that there are divergent links from national attachment, national pride, and national chauvinism to social and political trust. We do so with data from the US (General Social Survey) and the Netherlands (Longitudinal Internet Studies for the Social Sciences ), thus expanding current knowledge of national identity and trust to a highly relevant yet neglected European case.

Prioritizing professionals? How the democratic and professionalized nature of interest groups shapes their degree of access to EU officials [European Political Science Review]

Interest groups are key intermediary actors between civil society and public officials. The EU has long emphasized the importance of interacting with representative groups that involve their members. Additionally, there is an increasing trend toward the professionalization of groups that invest in organizational capacities to efficiently provide policy expertise. Both member involvement and organizational capacity are crucial features for groups to function as transmission belts that aggregate and transfer the preferences of their members to policymakers, thus reinforcing the legitimacy and efficiency of governance systems. Yet, not all groups have these organizational attributes. This paper quantitatively examines the effects of interest groups’ investment in member involvement and organizational capacity on the level of access to EU Commission officials. The results indicate that member involvement does not pay off in terms of higher levels of access. In contrast, groups with high organizational capacities have more meetings with public officials of the Commission.

Party decline or social transformation? Economic, institutional and sociological change and the rise of anti-political-establishment parties in Western Europe [European Political Science Review]

The rise in support for anti-political-establishment parties (APEp), especially since the beginning of the 2008 Great Recession, has put democracy in peril. Some scholars have warned us about the negative implications the recent rise of APEp might have for the development of democracy in Western Europe. For that reason, it is important we begin to understand what generates APEp’s electoral success. Drawing on a new comparative dataset that examines all Western European democracies from 1849 until 2017, the current article attempts to provide an explanation. In particular, our analyses examine three alternative explanations put forward by the literature: economic, institutional, and sociological. Our results show that it is not economic performance but both institutional and sociological change which together can help to understand the current wave of support for APEp.

Thursday, 24 September 2020

02:03 AM

Nutrition and the 2020 Pandemic [Annual Reviews: Annual Review of Nutrition: Table of Contents]

Annual Review of Nutrition, Volume 40, Issue 1, Page v-vi, September 2020.

Calorie Restriction and Aging in Humans [Annual Reviews: Annual Review of Nutrition: Table of Contents]

Annual Review of Nutrition, Volume 40, Issue 1, Page 105-133, September 2020.

Nutritional Requirements for Sustaining Health and Performance During Exposure to Extreme Environments [Annual Reviews: Annual Review of Nutrition: Table of Contents]

Annual Review of Nutrition, Volume 40, Issue 1, Page 221-245, September 2020.

Vitamin A and Retinoic Acid in Cognition and Cognitive Disease [Annual Reviews: Annual Review of Nutrition: Table of Contents]

Annual Review of Nutrition, Volume 40, Issue 1, Page 247-272, September 2020.

Drinking Water in the United States: Implications of Water Safety, Access, and Consumption [Annual Reviews: Annual Review of Nutrition: Table of Contents]

Annual Review of Nutrition, Volume 40, Issue 1, Page 345-373, September 2020.

Tuesday, 01 September 2020

02:04 AM

Copula-Based Analysis of Physical Layer Security Performances Over Correlated Rayleigh Fading Channels [IEEE Transactions on Information Forensics and Security - new TOC]

This paper studies performance analysis of physical layer security in a Rayleigh fading wiretap channel, where, the main (transmitter-to-legitimate receiver) and eavesdropper (transmitter-to-eavesdropper) channel coefficients are correlated. By exploiting the novel approach called Copula theory, we derive closed-form expressions for average secrecy capacity (ASC), secrecy outage probability (SOP), and secrecy coverage region (SCR). Moreover, to more evaluate the impact of channel correlation, the asymptotic behavior of SOP in high signal-to-noise ratio (SNR) regime and other scenarios is studied, and finally, the analytical results are illustrated numerically under the positive dependence, independence, and negative dependence structures. Based on the insights from this analysis, we found that the effect of correlated fading on the performance of physical layer security can be helpful or harmful in different scenarios, depending on the structure of dependency between the main and eavesdropper channels.

Tuesday, 04 August 2020

02:02 AM

How Distortions Alter the Impacts of International Trade in Developing Countries [Annual Reviews: Annual Review of Economics: Table of Contents]

Annual Review of Economics, Volume 12, Issue 1, Page 213-238, August 2020.

Robust Decision Theory and Econometrics [Annual Reviews: Annual Review of Economics: Table of Contents]

Annual Review of Economics, Volume 12, Issue 1, Page 239-271, August 2020.

Social Identity and Economic Policy [Annual Reviews: Annual Review of Economics: Table of Contents]

Annual Review of Economics, Volume 12, Issue 1, Page 355-389, August 2020.

Empirical Models of Lobbying [Annual Reviews: Annual Review of Economics: Table of Contents]

Annual Review of Economics, Volume 12, Issue 1, Page 391-413, August 2020.

Political Effects of the Internet and Social Media [Annual Reviews: Annual Review of Economics: Table of Contents]

Annual Review of Economics, Volume 12, Issue 1, Page 415-438, August 2020.

Friday, 31 July 2020

02:02 AM

Advances in the Science of Asking Questions [Annual Reviews: Annual Review of Sociology: Table of Contents]

Annual Review of Sociology, Volume 46, Issue 1, Page 37-60, July 2020.

Social Networks and Cognition [Annual Reviews: Annual Review of Sociology: Table of Contents]

Annual Review of Sociology, Volume 46, Issue 1, Page 159-174, July 2020.

Multiracial Categorization, Identity, and Policy in (Mixed) Racial Formations [Annual Reviews: Annual Review of Sociology: Table of Contents]

Annual Review of Sociology, Volume 46, Issue 1, Page 335-353, July 2020.

Contemporary Social Movements in a Hybrid Media Environment [Annual Reviews: Annual Review of Sociology: Table of Contents]

Annual Review of Sociology, Volume 46, Issue 1, Page 443-465, July 2020.

Thursday, 30 July 2020

06:02 PM

Violence in Latin America: An Overview of Research and Issues [Annual Reviews: Annual Review of Sociology: Table of Contents]

Annual Review of Sociology, Volume 46, Issue 1, Page 693-706, July 2020.

Tuesday, 12 May 2020

02:02 AM

Understanding Multilateral Institutions in Easy and Hard Times [Annual Reviews: Annual Review of Political Science: Table of Contents]

Annual Review of Political Science, Volume 23, Issue 1, Page 1-18, May 2020.

Democratic Stability: A Long View [Annual Reviews: Annual Review of Political Science: Table of Contents]

Annual Review of Political Science, Volume 23, Issue 1, Page 59-75, May 2020.

Understanding the Role of Racism in Contemporary US Public Opinion [Annual Reviews: Annual Review of Political Science: Table of Contents]

Annual Review of Political Science, Volume 23, Issue 1, Page 153-169, May 2020.

Transnational Actors and Transnational Governance in Global Environmental Politics [Annual Reviews: Annual Review of Political Science: Table of Contents]

Annual Review of Political Science, Volume 23, Issue 1, Page 203-220, May 2020.

How International Actors Help Enforce Domestic Deals [Annual Reviews: Annual Review of Political Science: Table of Contents]

Annual Review of Political Science, Volume 23, Issue 1, Page 357-383, May 2020.

Tuesday, 11 February 2020

03:00 PM

New Approaches to Target Inflammation in Heart Failure: Harnessing Insights from Studies of Immune Cell Diversity [Annual Reviews: Annual Review of Physiology: Table of Contents]

Annual Review of Physiology, Volume 82, Issue 1, Page 1-20, February 2020.

Neuronal Mechanisms that Drive Organismal Aging Through the Lens of Perception [Annual Reviews: Annual Review of Physiology: Table of Contents]

Annual Review of Physiology, Volume 82, Issue 1, Page 227-249, February 2020.

BMP Signaling in Development, Stem Cells, and Diseases of the Gastrointestinal Tract [Annual Reviews: Annual Review of Physiology: Table of Contents]

Annual Review of Physiology, Volume 82, Issue 1, Page 251-273, February 2020.

Regulation and Effects of FGF23 in Chronic Kidney Disease [Annual Reviews: Annual Review of Physiology: Table of Contents]

Annual Review of Physiology, Volume 82, Issue 1, Page 365-390, February 2020.

Marrow Adipocytes: Origin, Structure, and Function [Annual Reviews: Annual Review of Physiology: Table of Contents]

Annual Review of Physiology, Volume 82, Issue 1, Page 461-484, February 2020.


FeedRSSLast fetchedNext fetched after
Annual Reviews: Annual Review of Economics: Table of Contents XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
Annual Reviews: Annual Review of Nutrition: Table of Contents XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
Annual Reviews: Annual Review of Physiology: Table of Contents XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
Annual Reviews: Annual Review of Political Science: Table of Contents XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
Annual Reviews: Annual Review of Sociology: Table of Contents XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
cs.CR updates on XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
Early Edition XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
EurekAlert! - Breaking News XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
European Political Science Review XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
IEEE Transactions on Image Processing - new TOC XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
IEEE Transactions on Industrial Electronics - new TOC XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
IEEE Transactions on Industrial Informatics - new TOC XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
IEEE Transactions on Information Forensics and Security - new TOC XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
International Journal of Sports Physiology and Performance XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
JAMA Current Issue XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
Latest BMJ Research XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
Nature - Issue - science feeds XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020
The Lancet XML 10:02 AM, Thursday, 26 November 2020 01:02 PM, Thursday, 26 November 2020