ID 原文 译文
57838 在统一的编码器—交互与推理—输出框架下对此 4 类任务的已有研究进行了综述，并描述了2 种对此框架的可能扩展; Previous work on these four different types of subtasks is in- vestigated under a unified framework of encoder-interaction and reasoning-output. Two types of recent de- velopments for the frame are also given.
57839 最后讨论了机器阅读理解未来需要解决的问题. Several challenges on MＲC for future work are discussed at the end of the survey.
57840 为提高多接入边缘计算( MEC) 任务卸载效率，提出了一个任务卸载和异构资源调度的联合优化模型.考虑异构的通信资源和计算资源，联合最小化用户的设备能耗、任务执行时延和付费，并利用深度强化学习( DＲL) 算法对该模型求最优的任务卸载算法. In order to improve the task offloading efficiency in multi-access edge computing ( MEC) ，a joint optimization model for task offloading and heterogeneous resource scheduling was proposed，consid- ering the heterogeneous communication resources and computing resources，jointly minimizing the energy consumption of user equipment，task execution delay，and the payment. A deep reinforcement learning method is adopted in the model to obtain the optimal offloading algorithm.
57841 仿真结果表明，该优化算法比银行家算法的设备能耗、时延和付费的综合指标提升了 27.6% . Simulations show that the pro- posed algorithm improves the comprehensive indexes of equipment energy consumption，delay，and pay- ment by 27. 6% ，compared to the Banker's algorithm.
57842 为了实现 X-射线图片的焊接缺陷检测，采用了基于目标检测领域的经典模型———Faster Ｒ-CNN 的目标检测方法. Based on faster region-based convolutional neural networks ( Ｒ-CNN) model，a classical model in the field of object detection is used to achieve welding defect detection of X-ray images.
57843 WDXI 数据集是从大量的 X-射线图像整理和分类构建获得的，包括 7 种缺陷类型和无缺陷类型. A greatnumber of X-ray images are collected and sorted out into the dataset，called WDXI，including no-defecttype and 7 defect types.
57844 为了有效地提取焊接区域，提出了一种根据平均灰度值和单位面积内平均对比度值的改进方法. Firstly，an improved method can be used to extract the welding area effectivelyaccording to the average gray value and the average contrast value per unit area.
57845 经过实验验证，可采取自适应的直方图均衡化以及两次中值滤波的方法分别进行图像增强和降噪处理. The adaptive histogramequalization is used for image enhancement and double median blur is used for noise reduction after experimental comparison.
57846 最终，在焊接缺陷识别的多分类任务中，训练模型在测试集上达到了预期效果，不仅证明了 WDXI 数据集的研究价值，还为实现焊接缺陷的自动识别和定位进行了实验性的尝试. Finally，the testing on the pre-trained model is expected and acceptable in themulti-classification problem of welding defects recognition，and not only proves the research value ofWDXI，but also contributes towards making an experiment attempt for improving automatic classificationand localization of welding defects combined with Faster Ｒ-CNN model.
57847 为了实现 Scratch 可视化编程领域的作品分类，提出了一种基于标签关联性的多标签分类算法( MLLＲ) ，构建了一个有效的多标签 Scratch 分类模型. In order to implement the classification of projects in visual programming field of Scratch，a multi-label classification algorithm ( MLLＲ) appears based on label relevance. An effective multi-label classification model for Scratch projects was constructed.