Iterative Compilation Optimization Based on Metric Learning and Collaborative Filtering

Pass selection and phase ordering are two critical compiler auto-tuning problems. Traditional heuristic methods cannot effectively address these NP-hard problems especially given the increasing number of compiler passes and diverse hardware architectures. Recent research efforts have attempted to address these problems through machine learning. However, the large search space of candidate pass sequences, the large numbers of redundant and irrelevant features, and the lack of training program instances make it difficult to learn models well. Several methods have tried to use expert knowledge to simplify the problems, such as using only the compiler passes or subsequences in the standard levels (e.g., -O1, -O2, and -O3) provided by compiler designers. However, these methods ignore other useful compiler passes that are not contained in the standard levels. Principal component analysis (PCA) and exploratory factor analysis (EFA) have been utilized to reduce the redundancy of feature data. However, these unsupervised methods retain all the information irrelevant to the performance of compilation optimization, which may mislead the subsequent model learning. To solve these problems, we propose a compiler pass selection and phase ordering approach, called Iterative Compilation based on Metric learning and Collaborative filtering (ICMC) . First, we propose a data-driven method to construct pass subsequences according to the observed collaborative interactions and dependency among passes on a given program set. Therefore, we can make use of all available compiler passes and prune the search space. Then, a supervised metric learning method is utilized to retain useful feature information for compilation optimization while removing both the irrelevant and the redundant information. Based on the learned similarity metric, a neighborhood-based collaborative filtering method is employed to iteratively recommend a few superior compiler passes for each target program. Last, an iterative data enhancement method is designed to alleviate the problem of lacking training program instances and to enhance the performance of iterative pass recommendations. The experimental results using the LLVM compiler on all 32 cBench programs show the following: (1) ICMC significantly outperforms several state-of-the-art compiler phase ordering methods, (2) it performs the same or better than the standard level -O3 on all the test programs, and (3) it can reach an average performance speedup of 1.20 (up to 1.46) compared with the standard level -O3.

Managing Density Stress to Close the Maize Yield Gap

Continued yield increases of maize (Zea mays L.) will require higher planting populations, and enhancement of other agronomic inputs could alleviate density-induced stress. Row spacing, plant population, P-S-Zn fertility, K-B fertility, N fertility, and foliar protection were evaluated for their individual and cumulative impacts on the productivity of maize in a maize-soybean [Glycine max (L.) Merr.] rotation. An incomplete factorial design with these agronomic factors in both 0.76 and 0.51 m row widths was implemented for 13 trials in Illinois, United States, from 2014 to 2018. The agronomic treatments were compared to two controls: enhanced and standard, comprising all the factors applied at the enhanced or standard level, respectively. The 0.51 m enhanced management control yielded 3.3 Mg ha–1 (1.8–4.6 Mg ha–1 across the environments) more grain (25%) than the 0.76 m standard management control, demonstrating the apparent yield gap between traditional farm practices and attainable yield through enhanced agronomic management. Narrow rows and the combination of P-S-Zn and K-B fertility were the factors that provided the most significant yield increases over the standard control. Increasing plant population from 79,000 to 109,000 plants ha–1 reduced the yield gap when all other inputs were applied at the enhanced level. However, increasing plant population alone did not increase yield when no other factors were enhanced. Some agronomic factors, such as narrow rows and availability of plant nutrition, become more critical with increasing plant population when density-induced stress is more significant. Changes in yield were dependent upon changes in kernel number. Kernel weight was the heaviest when all the management factors were applied at the enhanced level while only planting 79,000 plants ha–1. Conversely, kernel weight was the lightest when increasing population to 109,000 plants ha–1 while all other factors were applied at the standard level. The yield contribution of each factor was generally greater when applied in combination with all other enhanced factors than when added individually to the standard input system. Additionally, the full value of high-input agronomic management was only realized when matched with greater plant density.

Assessing contribution of bottled water in nutrient absorption using the bottled water nutritional quality index (BWNQI) in Iran

In this study, the contribution of bottled water in the absorption of nutritional minerals in Iran has been investigated. To calculate the nutritional quality index of bottled water (BWNQI) and evaluate the contribution of bottled water in nutrient absorption; the concentration of nutrient minerals, the standard level of these elements in bottled water, the recommended amount of nutrient mineral and the total consumption of drinking water in different age-sex groups were analyzed. The results showed that the average contribution of bottled water in absorbing the recommended amount of the nutrients of fluoride (F), magnesium (Mg), calcium (Ca), sodium (Na), copper (Cu), zinc (Zn) and manganese (Mn) was 12.16, 4.98, 4.85, 2.12, 0.49, 0.33 and 0.02%, respectively. According to the BWNQI index, the bottled water quality was as follows: 53.5% poor, 36.6% marginal, 7% fair, 2.81% good. Although most of the bottled water studied in this research were mineral water, a significant portion of them had poor nutritional quality, so the addition of minerals needed by the body through bottled water should be given more attention by the bottled water manufacturers and suppliers.

Economic and industrial assessment of common plum and cherry plum in the foothill zone of Adygea

The article contains materials of the research on the yield of common plum and cherry plum varieties. The most productive samples have been selected. The productivity of the selected varieties have been studied for four years, which includes the following characteristics: average productivity (kg/tree), average productivity from one cubic meter of a crown (kg/m3 ), average fruit weight (g), average productivity (% to the standard). The economic efficiency of common plum and cherry plum for a 5 years period has been studied. The following indicators have taken into account: yield (t/ha), product cost (rubles/ha), production costs (rubles/ha), sales profit (rubles/ha), product profitability (%). It has been found that the productivity of the studied varieties (kg/tree) is quite different. Common plum productivity varies from from 14,4 (Nectar) to 34,6 (Shamsi). That of cherry plum varies from 20 (Rioni) to 34,1 (Shuntukskaya 11). Mathematical processing of data on the yield of fruits of common plum and cherry plum has been carried out using the method of field experiment. Statistical processing of one-factor experiment has been carried out by the method of deviation from the average one according to the variant. 3 groups of plum and cherry plum varieties have been identified at the standard level: low-productive, medium-productive, high-productive. The varieties of cherry plum with high productivity from one cubic meter of crown, more than 2,0 kg/m3 , have been identified: early Niberdzhaevskaya (st), Most early, Shuntukskaya 9, Shuntukskaya 11, Klyukovka, large Nalchikskaya. These include the following varieties of common plum: Renklod Altana, early Kabardinskaya (st), Vascova, Arvita, Hungarian Italian, Anna Shpet (st). The studied varieties of plum and cherry plum are divided into groups according to fruit sizes: small-fruited, medium-fruited and large-fruited. It has been established that the economic efficiency of the production of common plum and cherry plum fruits mainly depends on the yield of the variety and the cost of production. The profitability of common plum varieties is from 59,7% to 130%, cherry plum – from 39,9% to 59,2%.Economic and industrial assessment of common plum and cherryplum in the foothill zone of Adygea.

An exceedance probability of the influence in the Building Code

Во всем мире на картах сейсмической опасности приводятся наиболее вероятные значения сейсмических воздействий. В соответствии с допустимым риском эти значения могут быть увеличены. Например, в международных нормах для расчета сейсмостойкости радиационно-опасных объектов наиболее вероятные значения ускорений увеличиваются на величину стандартного отклонения. В отечественных строительных нормах предполагается использование наиболее вероятных значений. Считается, что вследствие различных факторов расчеты воздействий сопровождаются некоторыми погрешностями, вследствие чего оценки воздействий могут быть завышены на 10, 5 и 1% для карт ОСР А, В и С соответственно. Эти оценки были существенно увеличены при округлении оценок интенсивности до целочисленных значений баллов. Однако в пояснительной записке это не было отражено. Поэтому вместо 10, 5 и 1% для карт ОСР А, В и С следуют значения 50, 45 и 41% соответственно. Поскольку баллы нельзя непосредственно использовать в инженерных расчетах, согласно строительным нормам, производится пересчет баллов в ускорения по шкале MSK-64. Однако, еще создатели шкалы MSK-64 b карт ОСР и шкалы MSK-64 указывали, что шкала занижает оценки ускорений примерно в полтора раза. Причина – в шкале MSK-64 приводятся не реальные, а фильтрованные значения. Во время создания этой шкалы основная масса зданий имела не более 5 этажей. Фильтрацией удалялись низкочастотные составляющие, неопасные для существующих зданий. Цель работы – оценить реальную вероятность превышения уровня воздействий при расчетах сейсмостойкости зданий и сооружений. Методы исследования – анализ процедур оценки расчетных воздействий с использованием карт ОСР и строительных норм. Результаты работы – показано, что нормативный уровень ускорений при расчетах сейсмостойкости более, чем в два раза ниже реальных значений All over the world, the most probable values of seismic impacts are shown on seismic hazard maps. According to the acceptable risk, these values can be increased. In domestic building codes, it is customary to use the most likely values. It is believed that due to various factors, the impact calculations are accompanied by some errors, as a result of which the impact estimates may be overstated by 10, 5 and 1% for the general SRF maps A, B and C, respectively. However, these estimates were significantly increased when rounding the intensity estimates to integer values of the scores. However, this was not reflected in the explanatory note. Since the points cannot be directly used in engineering calculations, according to the building codes, the points are converted into accelerations on the MSK-64 scale. However, even the creators of the OSR maps and the MSK-64 scale indicated that the scale underestimates the acceleration estimates by about one and a half times. The Aim of the workis to assess the real probability of exceeding the impact level when calculating the seismic resistance of buildings and structures. Research methods - analysis of procedures for assessing calculated impacts using SRF maps and building codes. Results of the work - it is shown that the standard level of accelerations in the calculations of seismic resistance is more than twice lower than the real values.

Evaluation of the spring soft wheat collection in the conditions of the central region of the non-chernozem zone of Russia

The results of a study of 12 varieties of spring soft wheat in the conditions of Central region of the Nonchernozem zone of Russia in 2019-2020 are presented in order to isolate the most valuable source material in terms of yield and grain quality. A comprehensive assessment of economically useful traits was carried out - yield, resistance to fungal diseases and lodging, protein and gluten content, an assessment of baking qualities was carried out. Also, all varieties of the spring soft wheat collection were evaluated by the index method, and an index evaluation was carried out. It was found that the Ekada 66 variety (613.4 g / m2) is significantly more productive in comparison with the standard Zlata variety (449.9 g / m2). The varieties Hercules, KVS Akvilon, Ekada 109, Tyumenskaya 25, Chernozemno-Uralskaya 2 and Sudarynya also showed high yields, which are at the standard level. It was revealed that the varieties Hercules, KVS Akvilon, Ekada 109 and Ekada 66 were characterized by complex resistance to lodging and the main fungal diseases against a natural infectious background. It was found that the varieties Heracles, Kinelskaya Otrada, KVS Akvilon, Ekada 113 and Tyumenskaya 25 have the highest values of the complex indices, therefore, these varieties are recommended to be used in further breeding for high yield and grain quality.

Novel moisturized and antimicrobial hand gel based on zinc-aminoclay and Opuntia humifusa extract

The high antimicrobial ability and low toxicity of zinc-aminoclay (ZnAC) are claimed in our previous reports. In this study, we formulate a novel hand gel based on ZnAC and Opuntia humifusa (O. humifusa) extract, which is a high moisturizing agent. The antimicrobial activity, cytotoxicity, moisturizing effect, and clinical skin irritation of the hand gel are evaluated. The hand gel with 0.5 wt.% ZnAC and 1.0 v/v% O. humifusa extract can kill more than 99% Escherichia coli (gram-negative bacteria) and Staphylococcus aureus (gram-positive bacteria) after 24 h. Toxicity evaluation shows that, the hand gel does not affect the viability of mammalian HaCaT cells. Additionally, skin moisture is increased by applying the hand gel while its viscosity is at the standard level of commercial products. The hand gel has a skin irritation index of 0.0 and is classified as a non-irritating product. We successfully formulated hand gel from ZnAC, glucomannan, glycerol, and O. humifusa extract. Owing to the high antimicrobial activity and skin protection of hand gels, they are suitable to be used as hand sanitizers in restaurants, hospitals, and homes effectively.

Crossmodal Interactions Between Olfaction and Touch Affecting Well-Being and Perception of Cosmetic Creams

In the present series of studies, we investigated crossmodal perception of odor and texture. In four studies, participants tried two textures of face creams, one high viscosity (HV) and one low viscosity (LV), each with one of three levels of added odor (standard level, half of standard, or base [no added odor]), and then reported their levels of well-being. They also reported their perceptions of the face creams, including liking (global liking of the product, liking of its texture) and “objective” evaluations on just about right (JAR) scales (texture and visual appearance evaluations). In Study 1, women in France tried the creams on their hands, as they would when testing them in a store, and in Study 2, a second group of French women tried the creams on their faces, as they would at home. In Studies 3 and 4, these same two procedures were repeated in China. Results showed that both odor and texture had effects on well-being, liking, and JAR ratings, including interaction effects. Though effects varied by country and context (hand or face), the addition of odor to the creams generally increased reports of well-being, global liking and texture liking, in some cases affecting the “objective” evaluations of texture. This is one of the first investigations of crossmodal olfactory and tactile perception's impacts on well-being, and it reinforces previous literature showing the importance of olfaction on well-being.

Patterns of relationship between PM10 from air monitoring quality station and AOT data from MODIS sensor onboard of Terra satellite

Thailand, especially in the northern region, often encounters the problem of having PM10 exceeding the normal standard level, which could do harm to people’s health. Mostly, such problem is caused by the burning of forest area and open area; this is clearly seen during January–April of every year. Also, the problem as mentioned is caused by the meteorological conditions and the terrains in the northern region that make it easy for PM10 to be accumulated. The aim of this study was to analyze the patterns of relationship between PM10 measured from the ground monitoring station and AOT data received from MODIS sensor onboard of Terra satellite in Phrae Province located in the northern region of Thailand. The method performed was by analyzing the correlation between PM10 data obtained from the ground monitoring station and the AOT data received from the MODIS sensor onboard of Terra satellite during January–April 2018. It was found from the study that the change of the intensity of PM10 and AOT in the climate was highly related; it appeared that the correlation coefficient (r) in January–April was 0.92, 0.91, 0.91 and 0.92, respectively. This research pointed out that during February– –April, the areas of Phrae Province had the level of PM10 that affected health. Besides, from the method in this research, it revealed AOT data received from MODIS sensor onboard of Terra satellite could be applied in order to follow up, monitor, and notify the spatial changes of PM10 efficiently.

TO CALCULATION OF THE OPTIMAL DEGREE OF DAMAGE TO A BUILDING BY SEISMIC IMPACT

Дана постановка задачи по расчету оптимальной степени поврждения здания при сейсмическом воздействии. Условие оптимизационного расчета записано в виде вероятностно-экономической целевой функции, где начальные затраты на антисейсмическое усиление здания уравновешиваются с вероятностыми потерями, связанными с наступлением той или иной степени повреждения здания. Вероятность безотказности здания, соответствующая данному уравновешенному состоянию, названа оптимальной надежностью, а степень повреждения - оптимальной степенью повреждения. Всего в соответствии со шкалой MSK-64 рассмотрены 5 степеней повреждения. Последовательности переходов из одних состояний повреждения в другие представлены в виде простейшего Марковского потока с непрерывным временем и дискретными состояниями. Составлены дифференциальные уравнения процесса переходов и представлены их общие решения. В случае если последствия наступленияопределенной степени повреждения здания носят как экономический, так и неэкономическией характер, предложено, исходя из зависимости расчета экономических потерь и затрат от повышения надежности до высоких уровней и показателя нормативного уровня надежности здания, определенного по предлагаемому в литературе выражению, принять решения о допустимой степени повреждения здания. The problem statement is given for calculating the optimal degree of damage to a building under seismic impact. The condition of the optimization calculation is written in the form of a probabilistic-economic objective function, where the initial costs for antiseismic strengthening of the building are balanced with the probabilistic losses associated with the onset of one or another degree of damage to the building. The probability of a building's reliability, corresponding to a given balanced state, is called the optimal reliability, and the degree of damage is called the optimal degree of damage. In total, in accordance with the MSK-64 scale, 5 degrees of damage are considered. Sequences of transitions from one damage state to another are presented in the form of the simplest Markov flow with continuous time and discrete states. Differential equations of the transition process are compiled and their general solutions are presented. If the consequences of the onset of a certain degree of damage to the building are both economic and non-economic in nature, it is proposed, based on the dependence of the calculation of economic losses and costs from increasing reliability to high levels and the indicator of the standard level of building reliability, determined according to the expression proposed in the literature, to make decisions on permissible degree of damage to the building.