ID 原文 译文
25615 在计算全局和局部标记相关性时，我们使用了余弦相似性来获取不同标记之间的正相关性和负相关性，这样有助于我们进一步实现更可靠的多标记学习。 To calculating global and local label correlations, we utilize cosine similarity to obtain positive and negativec orrelations between different labels, which helps us to further achieve more reliable multi-label learning．
25616 我们在多种类型的数据集上进行了广泛的对比实验来验证所提算法的有效性。 We implemented extensive experimental comparisons based on various data sets to validate the effectiveness of our algorithm．
25617 实验结果表明，该算法显著优于大多数对比算法，展现出其在多标记学习中的突出性能。 The experimental results show that the proposed algorithm significantly outperforms most of the state-of-the art approaches, demonstrating its prominent performance for multi-label learning．
25618 网络表示学习是将网络中的节点映射到低维空间形成低维稠密特征向量的分布式学习方法。 Network representation learning is a distributed learning method that maps nodes in a network to low-dimensional spaces to form low-dimensional dense vectors．
25619 本文在现有网络表示学习研究的基础上，提出一种基于霍克斯点过程的动态网络表示学习方法。该方法基于霍克斯点过程有效结合了网络历史连边信息和网络演化中的三元闭包特性对当前节点产生连边的影响，解决了现有方法难以有效捕捉网络历史信息和演化特性的问题。 Based on the existing network representation learning research, this paper proposed a dynamic network representation learning method based on Hawkes point process, which effectively combines the network historical edges and the ternary closure characteristics in the network evolution to generate the new edges of the current nodes． It solves the problem that the existing methods are difficult to effectively capture the network historical information and evolution characteristics of dynamic networks．
25620 在多种数据集的实验结果表明，本文提出的方法较其它方法在节点分类、链路预测和可视化等实验中的性能均有较大的提高，实验中的 F1 分数值和 AUC 值分别提高了 3. 72% ～6. 41% 和 2. 22% ～4. 69% 。 Extensive experiments demonstrated that the embeddings learned from the proposed MHNE (Multivariate Hawkes process Network Embedding) model can achieve better performance than the state-of-the-art methods in downstream tasks, such as node classification and link prediction. The F1 score andAUC value in the experiments increased by 3. 72% ～ 6. 41% and 2. 22% ～ 4. 69% , respectively．
25621 针对海面通信受大气噪声干扰严重的问题，该文提出一种基于 DNN(Deep Neural Network)神经网络的单样本极化滤波器预测模型，研究其对海面短波地波通信链路中的大气噪声的抑制作用。 Aiming at the problem that the sea surface communication is seriously disturbed by atmospheric noise, this paper proposes a single sample polarization filter prediction model based on deep neural network, and studies its suppression effect on atmospheric noise in the sea surface short ground wave communication link．
25622 与传统算法不同，DNN 神经网络直接从大量输入数据获取信息间的非线性特性，并以此更新网络参数，通过对模型参数调整使得模型达到最优状态。 DNN neural network directly obtainsthe non-linear characteristics between information from a large amount of input data, which uses it to update the network parameters, and adjusts the model parameters to make the model reach the optimal state．
25623 选择三种脉冲成分比例不同的大气噪声进行仿真，结果表明传统算法与 DNN 网络模型在低信噪比约 0 ～ 15dB 时对信号的误码率影响基本一致，在高信噪比约超过 15dB，误码率达到 10－4时，深度学习模型比传统算法所需信噪比显著提高约5dB。 Three types of atmospheric noise with different proportions of pulse components are selected for simulation. The results show that the traditional algorithm and the DNN network model have basically the same effect on the signal error rate when the signal-to-noise ratio is about 0 ～ 15dB．When the bit error rate reaches 10－ 4, the deep learning model improves the signal-to-noise ratio by about 5dB compared with the traditional algorithm.
25624 实验结果验证了神经网络在单样本极化滤波器预测方向的可行性与准确性，具有很好的实用参考价值。 The experimental results verify the feasibility and accuracy of the neural network in predicting the direction of the single-sample polarization filter coefficients, which has good practical value．