With the complication of the subjects and environment of the machine learning, feature selection methods have been used more frequently as an effective mean of dimension reduction. However, existing ...feature selection methods are deficient in striking a balance between the relevance evaluation accuracy with the searching efficiency. In this regard, the characteristics of the relevance between the feature set and the classification result are analyzed. Then, we propose our Relevance Assignation Feature Selection (RAFS) method based on the mutual information theory, which assigns the relevance evaluation according to the redundancy. With this method, we can estimate the contribution of each feature in a feature set, which is regarded as value of the feature and is used as the heuristic index in searching of the relevant features. A special dataset (“Grid World”) with strong interactive features is designed. Using the Grid World and six other natural datasets, the proposed method is compared with six other feature selection methods. Results show that in the Grid World dataset, the RAFS method can find correct relevant features with the probability above 90%, much higher than the others. In six other datasets, the RAFS method also has the best performance in the classification accuracy.
A two-step severe shot peening (SSP) method was designed to investigate the softening mechanism of nanostructured Mg-8Gd-3Y (GW83) alloys and its effects on the surface characteristics of the ...deformation layer. The experimental results indicated that a gradient deformation layer with nanocrystalline (50–100 nm) near the surface layer was fabricated by single SSP (0.54 mmN). The corresponding formation mechanism of nanocrystalline was mainly owing to continuous dynamic recrystallization (cDRX). By applying the two-step SSP (0.54 + 0.23 mmN) treatment, softening occurred near the surface layer (<50 μm) due to the occurrence of partial discontinuous dynamic recrystallization (dDRX). The reason for dDRX behavior during SSP treatment resulted from the combined effects of high-level stored strain energy and heat production. This softening behavior was characterized by the increase of domain size, the slight enhancement of basal texture, and the decrease of dislocation density and microhardness. Therefore, appropriate SP parameters should be selected to avoid material softening when obtaining nanocrystalline Mg alloys.
•SSP developed nanostructured deformation layer and induced high-level CRS.•Two-step SSP method was designed to investigate the work softening mechanism in nanostructured GW83 alloy.•Partial dDRX behavior should be responsible for the softening phenomenon.•The softening phenomenon is characterized by the increase of domain size, the slight enhancement of basal texture, and the decrease of dislocation density.
With the rise of latency-sensitive and computationally intensive applications in mobile edge computing (MEC) environments, the computation offloading strategy has been widely studied to meet the ...low-latency demands of these applications. However, the uncertainty of various tasks and the time-varying conditions of wireless networks make it difficult for mobile devices to make efficient decisions. The existing methods also face the problems of long-delay decisions and user data privacy disclosures. In this paper, we present the FDRT, a federated learning and deep reinforcement learning-based method with two types of agents for computation offload, to minimize the system latency. FDRT uses a multi-agent collaborative computation offloading strategy, namely, DRT. DRT divides the offloading decision into whether to compute tasks locally and whether to offload tasks to MEC servers. The designed DDQN agent considers the task information, its own resources, and the network status conditions of mobile devices, and the designed D3QN agent considers these conditions of all MEC servers in the collaborative cloud-side end MEC system; both jointly learn the optimal decision. FDRT also applies federated learning to reduce communication overhead and optimize the model training of DRT by designing a new parameter aggregation method, while protecting user data privacy. The simulation results showed that DRT effectively reduced the average task execution delay by up to 50% compared with several baselines and state-of-the-art offloading strategies. FRDT also accelerates the convergence rate of multi-agent training and reduces the training time of DRT by 61.7%.
With the rapid development of cloud computing, there are more and more large-scale data centers, which makes the energy management of data centers more complex. In order to achieve better ...energy-saving effect, it is necessary to solve the problems of concurrent management and interdependence of IT, refrigeration, storage, and network equipment. Reinforcement learning learns by interacting with the environment, which is a good way to realize the independent management of the data center. In this paper, a overall energy consumption method for data center based on deep reinforcement learning is proposed to achieve collaborative energy saving of data center task scheduling and refrigeration equipment. A new multi-agent architecture is proposed to separate the training process from the execution process, simplify the interaction process during system operation and improve the operation effect. In the deep learning stage, a hybrid deep Q network algorithm is proposed to optimize the joint action value function of the data center and obtain the optimal strategy. Experiments show that compared with other reinforcement learning methods, our method can not only reduce the energy consumption of the data center, but also reduce the frequency of hot spots.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
Display omitted
•A correlation was found among surface residual stress, crack length, and the relaxation of residual stress fields.•The proposed assessment method achieved accurate prediction of ...fatigue crack growth rates.
The compressive residual stress field (CRSF) relaxes during the fatigue crack growth (FCG) process, and the pattern of this relaxation remains unclear. Residual stress (RS) relaxation leads to a change in residual stress intensity factor (RSIF), which significantly affects the fatigue crack growth rate (FCGR) and thus affects the assessment of fatigue crack growth rate. To this end, this work designed a series of experiments to explore the FCG behaviour in compressive residual stress fields (CRSFs) by considering RS relaxation. Firstly, shot peening(SP) treatments were conducted on the specimens. Then, residual stress fields (RSFs) at different crack propagation stages were measured to analyse the relationship between surface RS and inner RSF distribution. The derived weight function and the real-time RSF are combined to calculate the RSIF with/without considering RS relaxation. Finally, the effect of RS relaxation on the accuracy of FCGR assessment was investigated. The results show that the positive effect of CRS on the fatigue performance of the peened specimens is primarily embodied in the early stage of the FCG process. With the continuous relaxation of the CRSF, the difference in FCGR between the peened and unpeened specimens gradually decreases and tends to disappear. When the applied SIFs are large enough, the CRS would completely relax, and the inhibiting effect of CRSF on FCGR would fail. The proposed method in this work can effectively predict the FCGR of the peened specimen when considering the RS relaxation. This work could provide some meaningful insights for the assessment method of FCGR prediction in CRSF induced by surface strengthening treatment.
Background Postoperative delirium (POD) is a common anesthetic side effect in cardiac surgery. However, the role of oxygen saturation monitoring in reducing postoperative delirium has been ...controversial. Therefore, this meta-analysis aimed to analyze whether NIRS monitoring during cardiac surgery under cardiopulmonary bypass could reduce the incidence of postoperative delirium. Methods PubMed, Web of Science, Cochrane Library, Embase and China National Knowledge Infrastructure (CNKI) databases were systematically searched using the related keywords for randomized-controlled trials (RCTs) published from their inception to March 16, 2024. This review was conducted by the Preferred Reporting Project and Meta-Analysis Statement (PRISMA) guidelines for systematic review. The primary outcome was postoperative delirium, and the second outcomes included the length of ICU stay, the incidence of kidney-related adverse outcomes, and the incidence of cardiac-related adverse outcomes. Results The incidence of postoperative delirium could be reduced under the guidance of near-infrared spectroscopy monitoring (OR, 0.657; 95% CI, 0.447–0.965; P = 0.032; I 2 = 0%). However, there were no significant differences in the length of ICU stay (SMD, 0.005 days; 95% CI, −0.135–0.146; P = 0.940; I 2 = 39.3%), the incidence of kidney-related adverse outcomes (OR, 0.761; 95% CI, 0.386–1.500; P = 0.430; I 2 = 0%), and the incidence of the cardiac-related adverse outcomes (OR, 1.165; 95% CI, 0.556–2.442; P = 0.686; I 2 = 0%) between the two groups. Conclusion Near-infrared spectroscopy monitoring in cardiac surgery with cardiopulmonary bypass helps reduce postoperative delirium in patients. Systematic Review Registration PROSPERO, identifier, CRD42023482675
Research has established that the incorporation of 3D-printed lattice structures in cement substrates enhances the mechanical properties of cementitious materials. However, given that 3D-printing ...materials, notably polymers, exhibit varying degrees of mechanical performance under high-temperature conditions, their efficacy is compromised. Notably, at temperatures reaching 150 °C, these materials soften and lose their load-bearing capacity, necessitating further investigation into their compressive mechanical behavior in such environments. This study evaluates the compressibility of cement materials reinforced with lattice structures made from polyamide 6 (PA6) across different structural configurations and ambient temperatures, employing ABAQUS for simulation. Six distinct 3D-printed lattice designs with equivalent volume but varying configurations were tested under ambient temperatures of 20 °C, 50 °C, and 100 °C to assess their impact on compressive properties. The findings indicate that heightened ambient temperatures significantly diminish the reinforcing effect of 3D-printed materials on the properties of cement-based composites.
The automatic glioma segmentation is of great significance for clinical practice. This study aims to propose an automatic method based on superpixel for glioma segmentation from the T2 weighted ...Magnetic Resonance Imaging.
The proposed method mainly includes three steps. First, we propose an adaptive superpixel generation algorithm based on simple linear iterative clustering version with 0 parameter (ASLIC0). This algorithm can acquire a superpixel image with fewer superpixels and better fit the boundary of region of interest (ROI) by automatically selecting the optimal number of superpixels. Second, we compose a training set by calculating the statistical, texture, curvature and fractal features for each superpixel. Third, Support Vector Machine (SVM) is used to train classification model based on the features of the second step.
The experimental results on Multimodal Brain Tumor Image Segmentation Benchmark 2017 (BraTS2017) show that the proposed method has good segmentation performance. The average Dice, Hausdorff distance, sensitivity, and specificity for the segmented tumor against the ground truth are 0.8492, 3.4697 pixels, 81.47, and 99.64%, respectively. The proposed method shows good stability on high- and low-grade glioma samples. Comparative experimental results show that the proposed method has superior performance.
This provides a close match to expert delineation across all grades of glioma, leading to a fast and reproducible method of glioma segmentation.
Three novel low molecular weight polysaccharides (RLP-1a, RLP-2a, and RLP-3a) with 9004, 8761, and 7571 Da were first obtained by purifying the crude polysaccharides from the fruits of a traditional ...Chinese medicinal herb Rosae Laevigatae. The conditions for polysaccharides from the R. Laevigatae fruit (RLP) extraction were optimized by the response surface methodology, and the optimal conditions were as follows: extraction temperature, 93°C; extraction time, 2.8 h; water to raw material ratio, 22; extraction frequency, 3. Structural characterization showed that RLP-1a consisted of rhamnose, arabinose, xylose, glucose, and galactose with the ratio of 3.14 : 8.21 : 1 : 1.37 : 4.90, whereas RLP-2a was composed of rhamnose, mannose, glucose, and galactose with the ratio of 1.70 : 1 : 93.59 : 2.73, and RLP-3a was composed of rhamnose, arabinose, xylose, mannose, glucose, and galactose with the ratio of 6.04 : 26.51 : 2.05 : 1 : 3.17 : 31.77. The NMR analyses revealed that RLP-1a, RLP-2a, and RLP-3a contained 6, 4, and 6 types of glycosidic linkages, respectively. RLP-1a and RLP-3a exhibited distinct antioxidant abilities on the superoxide anions, 1,1-diphenyl-2-picrylhydrazyl (DPPH), and hydroxyl radicals in vitro. RLPs could decrease the serum lipid levels, elevate the serum high-density lipoprotein cholesterol levels, enhance the antioxidant enzymes levels, and upregulate of FADS2, ACOX3, and SCD-1 which involved in the lipid metabolic processes and oxidative stress in the high-fat diet-induced rats. These results suggested that RLPs ameliorated the high-fat diet- (HFD-) induced lipid metabolism disturbance in the rat liver through the peroxisome proliferator-activated receptor (PPAR) signaling pathway. Low molecular weight polysaccharides of RLP could be served as a novel potential functional food for improving hyperlipidemia and liver oxidative stress responses.
The effect of ultraviolet C (UV-C) radiation on the physicochemical properties and the bacterial diversity of fresh-cut cabbage was investigated over 12 days of storage at 5 °C. The color index (L*, ...a* and b*), decrease in weight, and content of soluble sugar, chlorophyll, ascorbic acid, total phenol, and malondialdehyde were measured. The structure of the bacterial community was also clarified by high-throughput sequencing analysis. After irradiation, the weight loss of fresh-cut cabbage during storage was reduced, and the content of antioxidant-related components, such as total phenols, ascorbic acid, and malondialdehyde, was increased. The results showed that the dominant phylum of bacteria in fresh-cut cabbage was Proteobacteria, and the dominant genus was Pseudomonas. A β-diversity analysis showed that the composition of the bacterial community of the irradiated and control treatments was different.