The purpose of unsupervised domain adaptation is to use the knowledge of the source domain whose data distribution is different from that of the target domain for promoting the learning task in the target *** key bott...
详细信息
The purpose of unsupervised domain adaptation is to use the knowledge of the source domain whose data distribution is different from that of the target domain for promoting the learning task in the target *** key bottleneck in unsupervised domain adaptation is how to obtain higher-level and more abstract feature representations between source and target domains which can bridge the chasm of domain ***,deep learning methods based on autoencoder have achieved sound performance in representation learning,and many dual or serial autoencoderbased methods take different characteristics of data into consideration for improving the effectiveness of unsupervised domain ***,most existing methods of autoencoders just serially connect the features generated by different autoencoders,which pose challenges for the discriminative representation learning and fail to find the real cross-domain *** address this problem,we propose a novel representation learning method based on an integrated autoencoders for unsupervised domain adaptation,called *** capture the inter-and inner-domain features of the raw data,two different autoencoders,which are the marginalized autoencoder with maximum mean discrepancy(mAE)and convolutional autoencoder(CAE)respectively,are proposed to learn different feature *** higher-level features are obtained by these two different autoencoders,a sparse autoencoder is introduced to compact these inter-and inner-domain *** addition,a whitening layer is embedded for features processed before the mAE to reduce redundant features inside a local *** results demonstrate the effectiveness of our proposed method compared with several state-of-the-art baseline methods.
The paper describes the details of using J-SIM in main memory database parallel recovery simulation. In update intensive main memory database systems, I/O is still the dominant performance bottleneck. A proposal of pa...
详细信息
The paper describes the details of using J-SIM in main memory database parallel recovery simulation. In update intensive main memory database systems, I/O is still the dominant performance bottleneck. A proposal of parallel recovery scheme for large-scale update intensive main memory database systems is presented. Simulation provides a faster way of evaluating the new idea compared to actual system implementation. J-SIM is an open source discrete time simulation software package. The simulation implementation using J-SIM is elaborated in terms of resource modeling, transaction processing system modeling and workload modeling. Finally, with simulation results analyzed, the effectiveness of the parallel recovery scheme is verified and the feasibility of J-SIM's application in main memory database system simulation is demonstrated.
Recent research has focused on density queries for moving objects in highly dynamic scenarios. An area is dense if the number of moving objects it contains is above some threshold. Monitoring dense areas has applicati...
详细信息
Domain adaptation aims to transfer knowledge from the labeled source domain to an unlabeled target domain that follows a similar but different ***,adversarial-based methods have achieved remarkable success due to the ...
详细信息
Domain adaptation aims to transfer knowledge from the labeled source domain to an unlabeled target domain that follows a similar but different ***,adversarial-based methods have achieved remarkable success due to the excellent performance of domain-invariant feature presentation ***,the adversarial methods learn the transferability at the expense of the discriminability in feature representation,leading to low generalization to the target *** this end,we propose a Multi-view Feature Learning method for the Over-penalty in Adversarial Domain ***,multi-view representation learning is proposed to enrich the discriminative information contained in domain-invariant feature representation,which will counter the over-penalty for discriminability in adversarial ***,the class distribution in the intra-domain is proposed to replace that in the inter-domain to capture more discriminative information in the learning of transferrable *** experiments show that our method can improve the discriminability while maintaining transferability and exceeds the most advanced methods in the domain adaptation benchmark datasets.
Construction of network-based academic resources is undergoing rapid development in this big data era, displaying a trend to replace digital libraries. This challenge calls for the construction of digital libraries to...
Construction of network-based academic resources is undergoing rapid development in this big data era, displaying a trend to replace digital libraries. This challenge calls for the construction of digital libraries to have big data related considerations, i.e., from resource construction aspect, we should expand the resource scope, aggrandize the breadth of resource integration, and increase the depth of the resource organization and processing;from technology application aspect, we should attach great importance to the semantic technology, emphasize the application of clustering technology, widely adopt data analysis technology, and elevate the retrieval technology level;from service aspect, we need to enrich services and product categories of digital libraries, and convert the passive, and general service mode to the proactive and personalized mode. The construction of next generation digital libraries must differ from the traditional way. Comprehensive service ideas should be established;comprehensive resource services should be provided;the conventional service mode and transmission form also need to be expanded;even the librarians of digital libraries are required to become data analysts.
The performance of online analytical processing （OLAP） is critical for meeting the increasing requirements of massive volume analytical applications. Typical techniques, such as in-memory processing, column-storage,...
详细信息
The performance of online analytical processing （OLAP） is critical for meeting the increasing requirements of massive volume analytical applications. Typical techniques, such as in-memory processing, column-storage, and join indexes focus on high perfor- mance storage media, efficient storage models, and reduced query processing. While they effectively perform OLAP applications, there is a vital limitation： main- memory database based OLAP （MMOLAP） cannot provide high performance for a large size data set. In this paper, we propose a novel memory dimension table model, in which the primary keys of the dimension table can be directly mapped to dimensional tuple addresses. To achieve higher performance of dimensional tuple access, we optimize our storage model for dimension tables based on OLAP query workload features. We present directly dimensional tuple accessing （DDTA） based join （DDTA- JOIN）, a technique to optimize query processing on the memory dimension table by direct dimensional tuple access. We also contribute by proposing an optimization of the predicate tree to shorten predicate operation length by pruning useless predicate processing. Our experimental results show that the DDTA-JOIN algorithm is superior to both simulated row-store main memory query processing and the open-source column-store main memory database MonetDB, thanks to the reduced join cost and simple yet efficient query processing.
Diagnosis of compound faults remains a challenge during fault diagnosis of bearings, owing to the different fault parameters coupling, fault characteristics diversity, and the exponential increasement of the number of...
详细信息
ISBN:
(纸本)9781665427470
Diagnosis of compound faults remains a challenge during fault diagnosis of bearings, owing to the different fault parameters coupling, fault characteristics diversity, and the exponential increasement of the number of possible failure modes. Current compound faults diagnostic methods, which are usually based on supervised or semi-supervised learning, require sufficient labeled or unlabeled training data for each compound faults. In industrial scenarios, neither labeled nor unlabeled training data of compound faults are usually difficult to collect and sometimes even inaccessible, whereas single faults samples are easy to obtain. Based on these issues, we construct a novel generative zero-shot learning (ZSL) compound faults diagnosis model identifies unseen compound faults using only single faults samples as training set. This model comprises several modules, namely semantic vector definition, feature extractor, generative adversarial modules. Firstly, we devise a unified semantic vector definition method for expressing single and compound faults based on theoretical correlation of characteristics between single fault and compound faults vibration data. Secondly, a CNN-based feature extractor is designed for extraction the fault features from the time-frequency domain of vibration data. Thirdly, a generative adversarial module performs adversarial training of semantic vectors and fault features of single faults to learn the mapping relationship between the fault features and the fault semantic vectors. Once trained, the generator is able to generate compound fault features using the compound fault semantic vectors, rather than any compound fault samples. Finally, the K-nearest neighbor method is adopted to identify the unseen compound faults by measuring the distance between the extracted feature from the testing compound fault samples and the generated features. The effectiveness of the proposed method is verified on a self-built bearing test stand. The results show
The remaining useful life (RUL) of bearings is critical to the proper operation of mechanical equipment, maintenance of equipment costs and availability. The existing domain adaptation methods have had great success i...
详细信息
Advances in wireless networks and positioning technologies (e.g., CPS) have enabled new data management applications that monitor moving objects. In such new applications, realtime data analysis such as clustering ana...
详细信息
ISBN:
(纸本)9783540717027
Advances in wireless networks and positioning technologies (e.g., CPS) have enabled new data management applications that monitor moving objects. In such new applications, realtime data analysis such as clustering analysis is becoming one of the most important requirements. In this paper, we present the problem of clustering moving objects in spatial networks and propose a unified framework to address this problem. Due to the innate feature of continuously changing positions of moving objects, the clustering results dynamically change. By exploiting the unique features of road networks, our framework first introduces a notion of cluster block (CB) as the underlying clustering unit. We then divide the clustering process into the continuous maintenance of CBs and periodical construction of clusters with different criteria based on CBs. The algorithms for efficiently maintaining and organizing the CBs to construct clusters are proposed. Extensive experimental results show that our clustering framework achieves high efficiency for clustering moving objects in real road networks.
In data management systems, query processing on GPUs or distributed clusters have proven to be an effective method for high efficiency. However, the high PCIe data transfer overhead between CPUs and GPUs, and the comm...
详细信息
暂无评论