Search

The problem of completing high-dimensional matrices from a limited set of observations arises in many big data applications, especially recommender systems. The existing matrix completion models generally follow either a memory- or a model-based approach, whereas geometric matrix completion (GMC) models combine the best from both approaches. Existing deep-learning-based geometric models yield good performance, but, in order to operate, they require a fixed structure graph capturing the relationships among the users and items. This graph is typically constructed by evaluating a pre-defined similarity metric on the available observations or by using side information, e.g., user profiles. In contrast, Markov-random-fields-based models do not require a fixed structure graph but rely on handcrafted features to make predictions. When no side information is available and the number of available observations becomes very low, existing solutions are pushed to their limits. In this article, we propose a GMC approach that addresses these challenges. We consider matrix completion as a structured prediction problem in a conditional random field (CRF), which is characterized by a maximum a posteriori (MAP) inference, and we propose a deep model that predicts the missing entries by solving the MAP inference problem. The proposed model simultaneously learns the similarities among matrix entries, computes the CRF potentials, and solves the inference problem. Its training is performed in an end-to-end manner, with a method to supervise the learning of entry similarities. Comprehensive experiments demonstrate the superior performance of the proposed model compared to various state-of-the-art models on popular benchmark data sets and underline its superior capacity to deal with highly incomplete matrices.

We report the karyotype, some aspects of spermatogenesis, and ovarian trophocytes ploidy in three aquatic bug species: Ilyocoris cimicoides (Linnaeus, 1758), Notonecta glauca Linnaeus, 1758, and Diplonychus rusticus Fabricius, 1871 from previously unexplored regions - South Europe (Bulgaria) and Southeast Asia (Vietnam). Our results add considerable support for the published karyotype data for these species. In I. cimicoides , we observed achiasmate male meiosis - the first report of achiasmy for the family Naucoridae. More comprehensive cytogenetic studies in other species of the Naucoridae are required to elucidate the role of achiasmy as a character in the systematics of the family. Our observations on the association between phases of spermatogenesis and developmental stages in I. cimicoides and N. glauca differ from the previously published data. In these species, we assume that the spermatogenesis phases are not strongly associated with certain developmental stages. For further cytogenetic studies (on the Balkan Peninsula), we recommend July as the most appropriate month for collection of I. cimicoides and N. glauca . In the ovaries of both species, we studied the level of ploidy in metaphase and interphase trophocytes. In I. cimicoides , diploid and tetraploid metaphase trophocytes were found. Heteropycnotic elements, observed in interphase trophocytes of this species, represented the X chromosomes. It allowed us to determine the trophocytes ploidy at interphase (2n was repeated up to 16 times). The situation with N. glauca was different. The metaphase trophocytes were diploid and we were not able to determine the ploidy of interphase trophocytes since such conspicuous heteropycnotic elements were not found. The scarce data available suggest a tendency for a low level of trophocyte ploidy in the basal infraorders (Nepomorpha and Gerromorpha) and for a high level in the more advanced Pentatomomorpha. Data about this character in species from other infraorders are needed to confirm that tendency., (Desislava Stoianova, Nikolay Simov, Manh Quang Vu, Duc Minh Nguyen, Snejana Grozeva.)

Background: HIV-infected children suffer from higher levels of treatment failure compared to adults. Immunoactivation, including humoral immunoactivation reflected by increased immunoglobulin levels, is believed to occur early during HIV infection. Therefore, we wanted investigate alteration in immunoglobulin levels in association with treatment response in HIV-infected children., Methods: A nested case-control study was conducted using clinical data collected from 68 HIV-infected children enrolled at the National Hospital of Pediatrics, Vietnam., Results: The results showed that immunoglobulin levels, CD4 T-cell counts, CD4 T-cell percentage, and HIV load were significantly higher in the treatment-failure group than the treatment-success group at treatment initiation. IgG and IgA levels were negatively correlated with CD4 T-cell counts ( P =0.049 and P <0.01, respectively) and positively correlated with HIV load ( P =0.04 and P =0.02, respectively). In addition, IgG and IgA levels were independently associated with treatment response, analyzed by Cox regression analysis (HR 1.19 [ P =0.049] and HR 1.69 [ P <0.01], respectively)., Conclusion: Elevation of IgA levels occurred early during HIV infection, and might have a prognostic role in treatment response., Competing Interests: Disclosure The authors report no conflicts of interest in this work.

Objectives: Over the past decades, Vietnam has made great strides in reducing the rate of mortality in HIV-related deaths, due to increased access of antiretroviral therapy (ART); however, given the significantly high level of treatment failure (TF), it is essential to identify markers that describe the failure of ART in HIV-1 infected children., Methods: A nested case-control study was conducted with clinical data collected from 101 HIV-infected children [26 TF and 75 treatment success (TS)] at National Hospital of Pediatrics, Vietnam (2008-2012)., Results: The results showed that certain factors including height, weight, vaccination with Hepatitis B, and platelet were significantly different between TF and TS before starting the treatment. In addition, age to start the treatment, CD4 percentage, and opportunistic infection were found to significantly predict treatment outcome most frequently, implying the importance of clinical markers in the treatment response by Cox regression analysis., Conclusions: There is an inherent complexity within clinical markers that is challenging to determine HIV-pediatric failure and further research is needed to build a complete picture to guide clinical, evidence-based practice.

Artificially structured materials with unit cells at sub-wavelength scale, known as metamaterials, have been widely used to precisely control and manipulate waves thanks to their unconventional properties which cannot be found in nature. In fact, the field of acoustic metamaterials has been much developed over the past 15 years and still keeps developing. Here, we present a topical review of metamaterials in acoustic wave science. Particular attention is given to fundamental principles of acoustic metamaterials for realizing the extraordinary acoustic properties such as negative, near-zero and approaching-infinity parameters. Realization of acoustic cloaking phenomenon which is invisible from incident sound waves is also introduced by various approaches. Finally, acoustic lenses are discussed not only for sub-diffraction imaging but also for applications based on gradient index (GRIN) lens.

A new xanthone, calothorexanthone, together with five known compounds, garbogiol, 1,4,8-trihydroxyxanthone, δ-tocotrienol, 1,7-dihydroxyxanthone and globuxanthone, was isolated from a petroleum ether extract of the bark of Calophyllum thorelii. Their structures were established on the basis of spectroscopic methods, mainly one- and two-dimensional NMR. Antioxidant activity of the isolated compounds was tested using DPPH free radical scavenging assay and some exhibited remarkable effects with IC₅₀ of 13.63-17.46 µg mL(-1).