摘要 :
Learning a fast global model that describes the observed phenomenon well is a crucial goal in the inherently distributed Vehicular Networks. This global model is further used for decision-making, which is especially important for ...
展开
Learning a fast global model that describes the observed phenomenon well is a crucial goal in the inherently distributed Vehicular Networks. This global model is further used for decision-making, which is especially important for some safety-related applications (i.e., the altering of accident and warning of traffic jam). Most existing works have ignored the network overhead caused by synchronizing with neighbors, which inevitably delays the time for agents to stabilize. In this paper, we focus on developing an asynchronous distributed clustering algorithm to learn the global model, where cluster models, rather than raw data points, are shared and updated. Empirical experiments on a message delay simulator show the efficiency of our methods, with a reduced convergence time, declined network overhead and improved accuracy (relative to the standard solution). This algorithm is further improved by introducing a tolerant delay. Compared to the algorithm without delay, the performance is improved significantly in terms of convergence time (by as much as 47%) and network overhead (by around 53%) if the underlying network is geometric or regular.
收起
摘要 :
It is a challenge to generate an accurate machine learning model in a distributed network due to the increased concern in data privacy and high cost in gathering all raw data. This paper presents an adaptive asynchronous distribut...
展开
It is a challenge to generate an accurate machine learning model in a distributed network due to the increased concern in data privacy and high cost in gathering all raw data. This paper presents an adaptive asynchronous distributed clustering algorithm and two centralised methods for agents in wireless network to learn the global models, while the privacy is protected. Moreover, the communication cost and clustering quality can be adaptively balanced. The proposed clustering algorithms do not require the number of clusters to be pre-defined, and we propose a bounding boxes based method to fully utilize the shape information of clusters to improve the accuracy of the global model. Furthermore, we consider different knowledge levels of agents and different requirements about the global model. In experiments on randomly generated network topologies, we demonstrate that methods which do all the iterations of clustering in each cycle, and which exchange descriptions of cluster shape and density instead of just centroids and data counts, achieve higher accuracy, in significantly shorter elapsed time.
收起
摘要 :
The relationship between baseline high peritoneal solute transport rate (PSTR) and the prognosis of peritoneal dialysis (PD) patients remains unclear. The present study combined clinical data and basic experiments to investigate t...
展开
The relationship between baseline high peritoneal solute transport rate (PSTR) and the prognosis of peritoneal dialysis (PD) patients remains unclear. The present study combined clinical data and basic experiments to investigate the impact of baseline PSTR and the underlying molecular mechanisms. A total of 204 incident CAPD patients from four PD centres in Shanghai between 1 January 2014 and 30 September 2020 were grouped based on a peritoneal equilibration test after the first month of dialysis. Analysed with multivariate Cox and logistic regression models, baseline high PSTR was a significant risk factor for technique failure (AHR 5.70; 95% CI 1.581 to 20.548 p =?0.008). Baseline hyperuricemia was an independent predictor of mortality (AHR 1.006 95%CI 1.003 to 1.008, p <?0.001) and baseline high PSTR (AOR 1.007; 95%CI 1.003 to 1.012; p =?0.020). Since uric acid was closely related to high PSTR and adverse prognosis, the in vitro experiments were performed to explore the underlying mechanisms of which uric acid affected peritoneum. We found hyperuricemia induced epithelial-to-mesenchymal transition (EMT) of cultured human peritoneal mesothelial cells by activating TGF-β1/Smad3 signalling pathway and nuclear transcription factors. Conclusively, high baseline PSTR induced by hyperuricaemia through EMT was an important reason of poor outcomes in CAPD patients.
收起
摘要 :
Due to the complex composition consisting of solid particles and fluids with different physical properties, geophysical flows often show complex and diverse dynamic characteristics. For landslides with high water content, there ar...
展开
Due to the complex composition consisting of solid particles and fluids with different physical properties, geophysical flows often show complex and diverse dynamic characteristics. For landslides with high water content, there are complex interactions between the solid and fluid phases. Therefore, it is difficult to grasp the dynamic characteristics and the disaster scale of this type of landslide, especially under complex terrain and ground conditions. The drag effect is an important aspect of the interaction between the solid and liquid phases. We optimized the enhanced drag coefficient formula to further consider the effect of high-velocity movement. By considering the volume fraction relationships between different phases, a mechanical erosion rate model is utilized for multiphase flows. Based on the r.avaflow numerical tool and the multiphase mass flow model, considering the interphase interaction characteristics of high-velocity liquefied landslides, we analyzed the influence of the obstruction of buildings and their entrainment into the landslide on the dynamic characteristics and hazard range of the Shenzhen 2015 landslide. This provides a reference for the analysis of complex geophysical disasters based on the multiphase mass flow model. Importantly, we have demonstrated the reduced mobility of the considered erosive impact event, which is in line with the physical principle.
收起
摘要 :
Wireless sensor networks, which are widely used in military, industrial and transportation fields, are vulnerable to various kinds of attacks, since they are mostly deployed in a relatively open environment. Based on the evolution...
展开
Wireless sensor networks, which are widely used in military, industrial and transportation fields, are vulnerable to various kinds of attacks, since they are mostly deployed in a relatively open environment. Based on the evolutionary game theory, this paper proposes a proactive defense model for wireless sensor networks, in which we emphasize that the node has a limited ability to learn the evolution of rationality from different attack strategies of the attacker, and can dynamically adjust their strategies to achieve the most effective defense. Following this approach, the cost (e.g., energy consumption and wastage of machinery) has been greatly saved and the life cycle of the nodes has been extended as well. By employing the proposed model, the whole wireless sensor network can be implemented in an effective way.
收起
摘要 :
Water is one of the most common and important objects on the earth, and its extraction is of great significance to many related researches in remote sensing domain. However, water always appears diversely, which makes its extracti...
展开
Water is one of the most common and important objects on the earth, and its extraction is of great significance to many related researches in remote sensing domain. However, water always appears diversely, which makes its extraction not so simple. Many former methods are developed to extract water, which mainly based on a single model and only use spectral information, but the results are not so satisfying. An adaptive extraction method based on normalized difference water index (NDWI) is proposed here to extract water completely and accurately from remote sensing image. This study first compute NDWI to enhance water’s spectral information, and then it is redefined so as to use the modified histogram auto-segmentation method to initially separate water from background; next, after segmentation, water pixels can be searched out and are taken as seed points to proceed region growing to get the local area of water; last, the edge of the local area is searched by a window template, and iterative classification within it is employed to precisely extract water’s precise partition. Experiments are carried out here on an ETM+ image of a paralic area to extract water. Through comparison with other commonly used methods, it shows that the performance of the proposed method is superior to the others.
收起
摘要 :
Water is one of the most common and important objects on the earth, and its extraction is of great significance to many related researches in remote sensing domain. However, water always appears diversely, which makes its extracti...
展开
Water is one of the most common and important objects on the earth, and its extraction is of great significance to many related researches in remote sensing domain. However, water always appears diversely, which makes its extraction not so simple. Many former methods are developed to extract water, which mainly based on a single model and only use spectral information, but the results are not so satisfying. An adaptive extraction method based on normalized difference water index (NDWI) is proposed here to extract water completely and accurately from remote sensing image. This study first compute NDWI to enhance water's spectral information, and then it is redefined so as to use the modified histogram auto-segmentation method to initially separate water from background; next, after segmentation, water pixels can be searched out and are taken as seed points to proceed region growing to get the local area of water; last, the edge of the local area is searched by a window template, and iterative classification within it is employed to precisely extract water's precise partition. Experiments are carried out here on an ETM+ image of a paralic area to extract water. Through comparison with other commonly used methods, it shows that the performance of the proposed method is superior to the others.
收起
摘要 :
Generating an accurate machine learning (ML) model is of great importance for the Internet of Vehicles (IoV). However, obtaining such a model is challenging due to the fact that subgroups of in-network vehicles receive data from d...
展开
Generating an accurate machine learning (ML) model is of great importance for the Internet of Vehicles (IoV). However, obtaining such a model is challenging due to the fact that subgroups of in-network vehicles receive data from different resources. A worthwhile investment then would be identifying those groups before inferring models. Similarity metrics are widely used to distinguish different groups. However, the efficiency of most existing similarity measurements is at the cost of increased computational complexity and decreased accuracy, making them unsuitable for IoV’s stringent conditions. To address this issue, we propose a computationally efficient method to measure the similarity of different vehicles, where a simplified version of Earth mover’s distance (EMD) is adopted. This distance metric is then embedded into a distributed clustering algorithm to learn the global pattern for vehicular systems. Our algorithm’s overall performance is measured using an asynchronous message delay simulator. Compared to the best algorithm of the state of the art, our proposed algorithm converges slightly slower (by less than 1%) but improves the clustering accuracy by as much as 20% with synthetic data. Additionally, real-world data collected from vehicles validates the efficiency of our proposed algorithm.
收起
摘要 :
Collaborative machine learning, especially Federated Learning (FL), is widely used to build high-quality Machine Learning (ML) models in the Internet of Vehicles (IoV). In this paper, we study the performance evaluation problem in...
展开
Collaborative machine learning, especially Federated Learning (FL), is widely used to build high-quality Machine Learning (ML) models in the Internet of Vehicles (IoV). In this paper, we study the performance evaluation problem in an inherently heterogeneous IoV, where the final models across the network are not identical and are computed on different standards. Previous studies assume that local agents are receiving data from the same phenomenon, and a same final model is fitted to them. However, this “one model fits all” approach leads to a biased performance evaluation of individual agents. We propose a general approach to measure the performance of individual agents, where the common knowledge and correlation between different agents are explored. Experimental results indicate that our evaluation scheme is efficient in these settings.
收起
摘要 :
Background In recent years, long noncoding RNAs (LncRNAs) have been found to play an important role in type 2 diabetes mellitus. However, research on the relationship between LncRNAs and prediabetes is still emerging. Objectives T...
展开
Background In recent years, long noncoding RNAs (LncRNAs) have been found to play an important role in type 2 diabetes mellitus. However, research on the relationship between LncRNAs and prediabetes is still emerging. Objectives The study aim was to screen differently expressed LncRNAs and understand their localization and function in patients with prediabetes. Methods We used microarray analysis to screen LncRNAs in prediabetes participants.To further clarify the localization and function of the expressed mRNAs, we used gene ontology analysis and pathway analysis. Then, internal validations were performed using individual quantitative real-time polymerase chain reaction (qRT-PCR) assays. Results We identified 55 differently expressed LncRNAs and 36 mRNAs in prediabetes participants comparing with controls. Gene ontology analysis indicated that the most enriched transcript terms were multicellular organismal process, plasma membrane, and binding. Pathway analysis indicated that the differently expressed mRNAs were involved in processes such as starch and sucrose metabolism, pantothenate and coenzyme A biosynthesis, and nicotinate and nicotinamide metabolism. The qRT-PCR results showed a trend consistent with the microarray results in 30 patients and 30 healthy controls. Conclusions We found aberrantly expressed LncRNAs and mRNAs in prediabetes subjects, and demonstrated that these LncRNAs are involved in the entire prediabetes biological process.
收起
原文获取