The Times has assembled a panel of distinguished and diverse writers who will regularly contribute to the Books section. 

The Kings let the Tampa Bay Lightning come back for a 4-2 win on the night Kobe Bryant is remembered.

When Kobe Bryant died in a helicopter crash, Ducks goalie John Gibson wanted to honor the Lakers legend with his mask. He knew just who to contact.

The Angels will pay at least $1.2 million to workers who would have staffed games at Angel Stadium, including those contracted by third parties.

Retailing for $2,070 and $3,150, respectively, the Philipp Plein tank top and hoodie are being sold to support the Mamba & Mambacita Sports Foundation, which was started by Lakers legend Kobe Bryant.

Kerry English was a beloved "professional audience member," a fan who saw up to five shows a week and provided the support theaters so critically need.

Appearing on "The Ellen DeGeneres Show," Grammys host Alicia Keys detailed the backstage frenzy that preceded her poignant tribute to Kobe Bryant.

Hollywood and Bollywood stars honor Indian actor Irrfan Khan ('Slumdog Millionaire,' 'Life of Pi'), who died Wednesday at 53.

Peter Whittingham's death has taken the world of football by shock.

THE commemoration was heartfelt, the nostalgia moving, the gratitude profound.

THE QUEEN's powerful VE Day speech touched the hearts of Britons around the country as she reflected on her personal memories of World War 2 and sent her "warmest good wishes" to everyone celebrating in lockdown.

THE commemoration was heartfelt, the nostalgia moving, the gratitude profound.

THE commemoration was heartfelt, the nostalgia moving, the gratitude profound.

F-35 jets fly over Utah as tribute to workers on frontline of COVID-19 fight

Officials would direct the Navy's Blue Angels and Air Force's Thunderbirds to fly over Indianapolis, Washington, New York and others

Indiana will issue unemployment benefits to workers who do not typically qualify. But distributing new federal stimulus money will take time.

Indianapolis rapper Tevin Studdard celebrates the return of Long's Bakery with "Long's Bakery Music Video," a tribute to glazed yeast doughnuts.

Node.js is used by many companies for building performant backend services without sacrificing developer efficiency. In this hands-on guide, author Thomas Hunter II proves that Node.js is just as capable as traditional enterprise platforms for building services that are observable, scalable, and resilient. Intermediate to advanced Node.js developers will find themselves integrating application code with a breadth of tooling from each layer of a modern service stack.

Interferon-regulated myxovirus resistance protein B (MxB) is an interferon-induced GTPase belonging to the dynamin superfamily. It inhibits infection with a wide range of different viruses, including HIV-1, by impairing viral DNA entry into the nucleus. Unlike the related antiviral GTPase MxA, MxB possesses an N-terminal region that contains a nuclear localization signal and is crucial for inhibiting HIV-1. Because MxB previously has been shown to reside in both the nuclear envelope and the cytoplasm, here we used bioinformatics and biochemical approaches to identify a nuclear export signal (NES) responsible for MxB's cytoplasmic location. Using the online computational tool LocNES (Locating Nuclear Export Signals or NESs), we identified five putative NES candidates in MxB and investigated whether their deletion caused nuclear localization of MxB. Our results revealed that none of the five deletion variants relocates to the nucleus, suggesting that these five predicted NES sequences do not confer NES activity. Interestingly, deletion of one sequence, encompassing amino acids 505–527, abrogated the anti-HIV-1 activity of MxB. Further mutation experiments disclosed that amino acids 515–519, and Pro-515 in particular, regulate MxB oligomerization and its binding to HIV-1 capsid, thereby playing an important role in MxB-mediated restriction of HIV-1 infection. In summary, our results indicate that none of the five predicted NES sequences in MxB appears to be required for its nuclear export. Our findings also reveal several residues in MxB, including Pro-515, critical for its oligomerization and anti-HIV-1 function.

Interferon-regulated myxovirus resistance protein B (MxB) is an interferon-induced GTPase belonging to the dynamin superfamily. It inhibits infection with a wide range of different viruses, including HIV-1, by impairing viral DNA entry into the nucleus. Unlike the related antiviral GTPase MxA, MxB possesses an N-terminal region that contains a nuclear localization signal and is crucial for inhibiting HIV-1. Because MxB previously has been shown to reside in both the nuclear envelope and the cytoplasm, here we used bioinformatics and biochemical approaches to identify a nuclear export signal (NES) responsible for MxB's cytoplasmic location. Using the online computational tool LocNES (Locating Nuclear Export Signals or NESs), we identified five putative NES candidates in MxB and investigated whether their deletion caused nuclear localization of MxB. Our results revealed that none of the five deletion variants relocates to the nucleus, suggesting that these five predicted NES sequences do not confer NES activity. Interestingly, deletion of one sequence, encompassing amino acids 505–527, abrogated the anti-HIV-1 activity of MxB. Further mutation experiments disclosed that amino acids 515–519, and Pro-515 in particular, regulate MxB oligomerization and its binding to HIV-1 capsid, thereby playing an important role in MxB-mediated restriction of HIV-1 infection. In summary, our results indicate that none of the five predicted NES sequences in MxB appears to be required for its nuclear export. Our findings also reveal several residues in MxB, including Pro-515, critical for its oligomerization and anti-HIV-1 function.

Heme-regulatory motifs (HRMs) are present in many proteins that are involved in diverse biological functions. The C-terminal tail region of human heme oxygenase-2 (HO2) contains two HRMs whose cysteine residues form a disulfide bond; when reduced, these cysteines are available to bind Fe3+-heme. Heme binding to the HRMs occurs independently of the HO2 catalytic active site in the core of the protein, where heme binds with high affinity and is degraded to biliverdin. Here, we describe the reversible, protein-mediated transfer of heme between the HRMs and the HO2 core. Using hydrogen-deuterium exchange (HDX)-MS to monitor the dynamics of HO2 with and without Fe3+-heme bound to the HRMs and to the core, we detected conformational changes in the catalytic core only in one state of the catalytic cycle—when Fe3+-heme is bound to the HRMs and the core is in the apo state. These conformational changes were consistent with transfer of heme between binding sites. Indeed, we observed that HRM-bound Fe3+-heme is transferred to the apo-core either upon independent expression of the core and of a construct spanning the HRM-containing tail or after a single turnover of heme at the core. Moreover, we observed transfer of heme from the core to the HRMs and equilibration of heme between the core and HRMs. We therefore propose an Fe3+-heme transfer model in which HRM-bound heme is readily transferred to the catalytic site for degradation to facilitate turnover but can also equilibrate between the sites to maintain heme homeostasis.

The Government will distribute free reusable face masks to all Hong Kong citizens, the Innovation & Technology Bureau announced today.

The CuMask, made with six layers and special ergonomic features, was developed by the Hong Kong Research Institute of Textiles & Apparel.

Two of its layers contain copper which is capable of immobilising bacteria, common viruses and other harmful substances.

The mask complies with the American Society for Testing & Materials F2100 Level 1 Standard in terms of particle and bacterial filtration efficiency, resistance to penetration by synthetic blood, and flammability and pressure resistance.

It is also reusable for up to 60 washes.

The bureau said, except for babies and infants, all holders of valid Hong Kong identity cards are eligible to obtain a mask.

Citizens can register online from 7am tomorrow till June 6. Each registration can cater for a maximum of six persons.

Upon successful registration, the mask will be delivered to the door by Hongkong Post within two weeks.

Primary and kindergarten students will each be given two masks, which will be delivered directly to children's schools. Parents do not have to register.

The Government has also arranged to deliver over 140,000 of the masks to residential homes and social welfare institutions for their distribution to those including elderly and the homeless.

Click here for registration details.

Heme-regulatory motifs (HRMs) are present in many proteins that are involved in diverse biological functions. The C-terminal tail region of human heme oxygenase-2 (HO2) contains two HRMs whose cysteine residues form a disulfide bond; when reduced, these cysteines are available to bind Fe3+-heme. Heme binding to the HRMs occurs independently of the HO2 catalytic active site in the core of the protein, where heme binds with high affinity and is degraded to biliverdin. Here, we describe the reversible, protein-mediated transfer of heme between the HRMs and the HO2 core. Using hydrogen-deuterium exchange (HDX)-MS to monitor the dynamics of HO2 with and without Fe3+-heme bound to the HRMs and to the core, we detected conformational changes in the catalytic core only in one state of the catalytic cycle—when Fe3+-heme is bound to the HRMs and the core is in the apo state. These conformational changes were consistent with transfer of heme between binding sites. Indeed, we observed that HRM-bound Fe3+-heme is transferred to the apo-core either upon independent expression of the core and of a construct spanning the HRM-containing tail or after a single turnover of heme at the core. Moreover, we observed transfer of heme from the core to the HRMs and equilibration of heme between the core and HRMs. We therefore propose an Fe3+-heme transfer model in which HRM-bound heme is readily transferred to the catalytic site for degradation to facilitate turnover but can also equilibrate between the sites to maintain heme homeostasis.

(Keck School of Medicine of USC) A new USC study suggests that temporarily suppressing the body's immune system during the early stages of COVID-19 could help a patient avoid severe symptoms. That's because the research shows that an interaction between the body's two main lines of defense may be causing the immune system to go into overdrive in some patients.

(Carnegie Mellon University) A new study sought to determine how the impact of recommender systems (also called recommenders) is affected by factors such as product type, attributes, and other sources of information about products on retailers' websites. The study found that recommenders increased the number of consumer views of product pages as well as the number of products consumers consider, but that the increase was moderated by product attributes and review ratings.

Members of the Jamaica Constabulary Force on Thursday, distributed more than 300 food packages to individuals and families impacted by COVID-19 in St. James, under the Private Sector Organisation of Jamaica (PSOJ)-led COVID-19 Response Fund...

Heme-regulatory motifs (HRMs) are present in many proteins that are involved in diverse biological functions. The C-terminal tail region of human heme oxygenase-2 (HO2) contains two HRMs whose cysteine residues form a disulfide bond; when reduced, these cysteines are available to bind Fe3+-heme. Heme binding to the HRMs occurs independently of the HO2 catalytic active site in the core of the protein, where heme binds with high affinity and is degraded to biliverdin. Here, we describe the reversible, protein-mediated transfer of heme between the HRMs and the HO2 core. Using hydrogen-deuterium exchange (HDX)-MS to monitor the dynamics of HO2 with and without Fe3+-heme bound to the HRMs and to the core, we detected conformational changes in the catalytic core only in one state of the catalytic cycle—when Fe3+-heme is bound to the HRMs and the core is in the apo state. These conformational changes were consistent with transfer of heme between binding sites. Indeed, we observed that HRM-bound Fe3+-heme is transferred to the apo-core either upon independent expression of the core and of a construct spanning the HRM-containing tail or after a single turnover of heme at the core. Moreover, we observed transfer of heme from the core to the HRMs and equilibration of heme between the core and HRMs. We therefore propose an Fe3+-heme transfer model in which HRM-bound heme is readily transferred to the catalytic site for degradation to facilitate turnover but can also equilibrate between the sites to maintain heme homeostasis.

Choreographer and designer Barry Moncrieffe; actress Lois Kelly Miller; New York-based broadcaster Gil Bailey; and actor and teacher Carl Samuels will all be remembered with a special tribute at Saturday’s staging of the annual Actor Boy Awards....

THE EDITOR, Madam: Words cannot describe my feeling of loss on hearing that prominent psychiatrist Professor Fred Hickling has died. Professor Hickling was one of Jamaica’s treasures through his work in the field of community mental health. His...

15 November 2016

During a time when the world is focused on the current COVID-19 pandemic, a physical health crisis, and all those treating the physical symptoms of this pandemic, we must not...

I have never watched Hidden Figures. Was that a bad way to start off an article concerning the late Katherine Johnson, the NASA legend whose persistence, precision, and proclivity for mathematics sent America to space in the 1960s? Maybe it was. … Continue reading →

[Rome], [1725]

[Rome], [1725]

Calcutta, India : Cleanair, Campaign for Smoke-free Environment, [198-?]

Feature screening is a powerful tool in processing high-dimensional data. When the sample size N and the number of features p are both large, the implementation of classic screening methods can be numerically challenging. In this paper, we propose a distributed screening framework for big data setup. In the spirit of 'divide-and-conquer', the proposed framework expresses a correlation measure as a function of several component parameters, each of which can be distributively estimated using a natural U-statistic from data segments. With the component estimates aggregated, we obtain a final correlation estimate that can be readily used for screening features. This framework enables distributed storage and parallel computing and thus is computationally attractive. Due to the unbiased distributive estimation of the component parameters, the final aggregated estimate achieves a high accuracy that is insensitive to the number of data segments m. Under mild conditions, we show that the aggregated correlation estimator is as efficient as the centralized estimator in terms of the probability convergence bound and the mean squared error rate; the corresponding screening procedure enjoys sure screening property for a wide range of correlation measures. The promising performances of the new method are supported by extensive numerical examples.

We propose graph-dependent implicit regularisation strategies for synchronised distributed stochastic subgradient descent (Distributed SGD) for convex problems in multi-agent learning. Under the standard assumptions of convexity, Lipschitz continuity, and smoothness, we establish statistical learning rates that retain, up to logarithmic terms, single-machine serial statistical guarantees through implicit regularisation (step size tuning and early stopping) with appropriate dependence on the graph topology. Our approach avoids the need for explicit regularisation in decentralised learning problems, such as adding constraints to the empirical risk minimisation rule. Particularly for distributed methods, the use of implicit regularisation allows the algorithm to remain simple, without projections or dual methods. To prove our results, we establish graph-independent generalisation bounds for Distributed SGD that match the single-machine serial SGD setting (using algorithmic stability), and we establish graph-dependent optimisation bounds that are of independent interest. We present numerical experiments to show that the qualitative nature of the upper bounds we derive can be representative of real behaviours.

In many areas, practitioners need to analyze large data sets that challenge conventional single-machine computing. To scale up data analysis, distributed and parallel computing approaches are increasingly needed. Here we study a fundamental and highly important problem in this area: How to do ridge regression in a distributed computing environment? Ridge regression is an extremely popular method for supervised learning, and has several optimality properties, thus it is important to study. We study one-shot methods that construct weighted combinations of ridge regression estimators computed on each machine. By analyzing the mean squared error in a high-dimensional random-effects model where each predictor has a small effect, we discover several new phenomena. Infinite-worker limit: The distributed estimator works well for very large numbers of machines, a phenomenon we call 'infinite-worker limit'. Optimal weights: The optimal weights for combining local estimators sum to more than unity, due to the downward bias of ridge. Thus, all averaging methods are suboptimal. We also propose a new Weighted ONe-shot DistributEd Ridge regression algorithm (WONDER). We test WONDER in simulation studies and using the Million Song Dataset as an example. There it can save at least 100x in computation time, while nearly preserving test accuracy.

When the data is distributed across multiple servers, lowering the communication cost between the servers (or workers) while solving the distributed learning problem is an important problem and is the focus of this paper. In particular, we propose a fast, and communication-efficient decentralized framework to solve the distributed machine learning (DML) problem. The proposed algorithm, Group Alternating Direction Method of Multipliers (GADMM) is based on the Alternating Direction Method of Multipliers (ADMM) framework. The key novelty in GADMM is that it solves the problem in a decentralized topology where at most half of the workers are competing for the limited communication resources at any given time. Moreover, each worker exchanges the locally trained model only with two neighboring workers, thereby training a global model with a lower amount of communication overhead in each exchange. We prove that GADMM converges to the optimal solution for convex loss functions, and numerically show that it converges faster and more communication-efficient than the state-of-the-art communication-efficient algorithms such as the Lazily Aggregated Gradient (LAG) and dual averaging, in linear and logistic regression tasks on synthetic and real datasets. Furthermore, we propose Dynamic GADMM (D-GADMM), a variant of GADMM, and prove its convergence under the time-varying network topology of the workers.

Jianqing Fan, Dong Wang, Kaizheng Wang, Ziwei Zhu.

Source: The Annals of Statistics, Volume 47, Number 6, 3009--3031.

Abstract:
Principal component analysis (PCA) is fundamental to statistical machine learning. It extracts latent principal factors that contribute to the most variation of the data. When data are stored across multiple machines, however, communication cost can prohibit the computation of PCA in a central location and distributed algorithms for PCA are thus needed. This paper proposes and studies a distributed PCA algorithm: each node machine computes the top $K$ eigenvectors and transmits them to the central server; the central server then aggregates the information from all the node machines and conducts a PCA based on the aggregated information. We investigate the bias and variance for the resulting distributed estimator of the top $K$ eigenvectors. In particular, we show that for distributions with symmetric innovation, the empirical top eigenspaces are unbiased, and hence the distributed PCA is “unbiased.” We derive the rate of convergence for distributed PCA estimators, which depends explicitly on the effective rank of covariance, eigengap, and the number of machines. We show that when the number of machines is not unreasonably large, the distributed PCA performs as well as the whole sample PCA, even without full access of whole data. The theoretical results are verified by an extensive simulation study. We also extend our analysis to the heterogeneous case where the population covariance matrices are different across local machines but share similar top eigenstructures.