In order to meet the performance requirements and reduce resource consumption, researchers have proposed many autoscaling schemes. However, most of them only considered current states of the servers or application tha...
详细信息
ISBN:
(纸本)9781538667521
In order to meet the performance requirements and reduce resource consumption, researchers have proposed many autoscaling schemes. However, most of them only considered current states of the servers or application that limits the applicability of autoscaler. This paper presents an application-oriented autoscaler based on Long Short Term Memory network and Multi-Layer perceptron. The autoscaler includes a workload prediction model, a response time prediction model and a resource adjusting model. With these models, workload and response time of service are accurately predicted, and mean absolute percentage error of workload prediction is reduced to 3.3 × 10 -4 . The autoscaler can also provide appropriate adjustment suggestion to server managers. The experiments compared our autoscaler with other models, and results show that our autoscaler had better performance in workload prediction and server adjusting.
Hyperspectral image classification faces various levels of difficulty due to the use of different types of hyperspectral image data. Recently, spectral-spatial approaches have been developed by jointly taking care of ...
详细信息
Many applications require learning classifiers or regressors that are both accurate and cheap to evaluate. Prediction cost can be drastically reduced if the learned predictor is constructed such that on the majority o...
ISBN:
(纸本)9781510860964
Many applications require learning classifiers or regressors that are both accurate and cheap to evaluate. Prediction cost can be drastically reduced if the learned predictor is constructed such that on the majority of the inputs, it uses cheap features and fast evaluations. The main challenge is to do so with little loss in accuracy. In this work we propose a budget-aware strategy based on deep boosted regression trees. In contrast to previous approaches to learning with cost penalties, our method can grow very deep trees that on average are nonetheless cheap to compute. We evaluate our method on a number of datasets and find that it outperforms the current state of the art by a large margin. Our algorithm is easy to implement and its learning time is comparable to that of the original gradient boosting. Source code is made available at http://***/svenpeter42/LightGBM-CEGB.
In the original version of this Article the values in the rightmost column of Table 1 were inadvertently shifted relative to the other columns. This has now been corrected in the PDF and HTML versions of the Article.
In the original version of this Article the values in the rightmost column of Table 1 were inadvertently shifted relative to the other columns. This has now been corrected in the PDF and HTML versions of the Article.
Although it has become an accepted lay view that when labeling objects through crowdsourcing systems, non-expert annotators often exhibit biases, this argument lacks sufficient evidential observation and systematic em...
Although it has become an accepted lay view that when labeling objects through crowdsourcing systems, non-expert annotators often exhibit biases, this argument lacks sufficient evidential observation and systematic empirical study. This paper initially analyzes eight real-world datasets from different domains whose class labels were collected from crowdsourcing systems. Our analyses show that biased labeling is a systematic tendency for binary categorization; in other words, for a large number of annotators, their labeling qualities on the negative class (supposed to be the majority) are significantly greater than are those on the positive class (minority). Therefore, the paper empirically studies the performance of four existing EM-based consensus algorithms , DS, GLAD, RY, and ZenCrowd, on these datasets. Our investigation shows that all of these state-of-the-art algorithms ignore the potential bias characteristics of datasets and perform badly although they model the complexity of the systems. To address the issue of handling biased labeling, the paper further proposes a novel consensus algorithm, namely adaptive weighted majority voting (AWMV), based on the statistical difference between the labeling qualities of the two classes. AWMV utilizes the frequency of positive labels in the multiple noisy label set of each example to obtain a bias rate and then assigns weights derived from the bias rate to negative and positive labels. Comparison results among the five consensus algorithms (AWMV and the four existing) show that the proposed AWMV algorithm has the best overall performance. Finally, this paper notes some potential related topics for future study.
Gliomas are the most common primary brain malignancies, with different degrees of aggressiveness, variable prognosis and various heterogeneous histologic sub-regions, i.e., peritumoral edematous/invaded tissue, necrot...
详细信息
Diffusion tensor imaging （DTI） is known to be the best non-invasive imaging modality in providing anatomical information as white-matter fiber bundles. However, the Gaussian noise introduced into the diffusion tenso...
详细信息
ISBN:
(纸本)9781467321969
Diffusion tensor imaging （DTI） is known to be the best non-invasive imaging modality in providing anatomical information as white-matter fiber bundles. However, the Gaussian noise introduced into the diffusion tensor images can bring serious impacts on tensor calculation and fiber tracking. To decrease the effects of the Gaussian noise, many denoising methods have been presented. In this paper, a shearlet based denosing strategy is introduced. To evaluate the efficiency of the proposed shearlet based denoising method in accounting for the Gaussian noise introduced into the images, the peak to peak signal-to-noise ratio （PSNR）, signal-to-mean squared error ratio （SMSE） and edge keeping index （Beta） metrics are adopted. The experiment results acquired from both the synthetic and real data indicate the good performance of our proposed filter.
暂无评论