ID	原文	译文
56948	基于开源的FAST平台的实验结果表明,与通用的BIND9系统相比响应延迟降低了约10倍,吞吐量接近10 Gb链路线速,同时资源开销小且具有良好的可扩展性.	The experimental results based onthe open-source FAST platform show that the response latency is reduced by approximately 10 times comparedwith the general BIND9 system, and the throughput is near the 10 Gb link-line speed. Moreover, the resourceoverhead is small, and the system is scalable.
56949	步态特征识别是生物特征识别的一种,在大量实际场景中有广泛的应用.	Gait recognition is a type of biometric recognition that can be used as an identification tool in variousapplications.
56950	目前,基于深度学习的方法在步态识别任务中表现出较好的准确率.但是,在对机器学习的研究中,人们不仅希望得到精确的预测,还希望算法对识别结果进行解释,以便人们理解实际问题中的关键.	Deep learning-based methods have recently exhibited promising accuracy in gait recognition tasks;however, in addition to an accurate prediction, these methods are required to explain the recognition results. The black-box nature of deep neural networks makes it very difficult to interpret the basis for their identification.
56951	深度神经网络的黑盒属性使得解释其识别依据非常困难.	The published studies on the interpretability of gait recognition are also in a blank state.
56952	在已有的步态识别文献中,关注可解释性的研究尚处于空白状态.另外,深度神经网络需要大量数据来学习模型参数,在问题规模较小时难以有效地在未见数据上泛化.	Moreover, deepneural networks require a large amount of data to learn the model parameters and an effective generalizationon unseen data is difficult when the problem size is small.
56953	本文探索了一种兼具准确性和可解释性的步态识别方法.	Thus, this paper presents a gait recognition methodcombining accuracy and interpretability. The gait feature is represented as a multi-dimensional time series and aShapelet-based time series classification method is used for gait recognition.
56954	将步态特征表示为多维时间序列,使用一种基于Shapelet的时间序列分类方法进行步态识别.	A Shapelet is the most discriminativesubsequence in time series that makes the proposed method provide interpretability and accuracy simultaneously.
56955	Shapelet是时间序列中最具有辨别性的子序列,基于Shapelet的时间序列分类方法能够提供较好的可解释性,同时可以提供较高的准确率.	We conducted experiments on the CASIA-B dataset and compared the proposed method with several state-of?the-arts deep learning methods.
56956	我们在CASIA-B数据集上进行了实验,和几种较新的深度学习方法进行了比较.	Experiments show that the proposed method can provide an accuracy close tothat of deep neural networks on small-scale data sets.
56957	实验表明,本文提出的方法在较小规模的数据集上能够提供与深度神经网络接近的准确率.	At the same time, the decision-making reason of the modelcan be explained in detail.