To improve the accuracy of short text matching,a short text matching method with knowledge and structure enhancement for BERT(KS-BERT)was proposed in this study.This method first introduced external knowledge to the i...To improve the accuracy of short text matching,a short text matching method with knowledge and structure enhancement for BERT(KS-BERT)was proposed in this study.This method first introduced external knowledge to the input text,and then sent the expanded text to both the context encoder BERT and the structure encoder GAT to capture the contextual relationship features and structural features of the input text.Finally,the match was determined based on the fusion result of the two features.Experiment results based on the public datasets BQ_corpus and LCQMC showed that KS-BERT outperforms advanced models such as ERNIE 2.0.This Study showed that knowledge enhancement and structure enhancement are two effective ways to improve BERT in short text matching.In BQ_corpus,ACC was improved by 0.2%and 0.3%,respectively,while in LCQMC,ACC was improved by 0.4%and 0.9%,respectively.展开更多
复杂电磁环境下卫星信号往往淹没在背景和噪声中,传统的信号检测算法在没有准确先验知识的情况下性能急剧降低,目前基于深度学习的信号检测算法往往需要依赖专家经验的数据后处理步骤,无法对信号进行端到端检测.针对上述缺陷,提出一种基...复杂电磁环境下卫星信号往往淹没在背景和噪声中,传统的信号检测算法在没有准确先验知识的情况下性能急剧降低,目前基于深度学习的信号检测算法往往需要依赖专家经验的数据后处理步骤,无法对信号进行端到端检测.针对上述缺陷,提出一种基于DETR_S(DEtection with TRansformer on Signal)的卫星信号智能检测方法.DETR_S以编码器-解码器架构为基础,利用Transformer网络全局建模能力捕获频谱信息,采用多头自注意力机制有效改善频谱信息长距离依赖的问题.基于匈牙利算法的预测框匹配模块摒弃了非极大值抑制的数据后处理步骤,将信号检测问题转变为集合预测问题,使模型并行输出检测结果.引入信号重构模块,将频谱重构损失函数加入损失函数中,辅助模型挖掘频谱深层表征,提升信号检测性能.实验结果表明,在仅使用信号频谱幅度信息条件下,DETR_S能够在信噪比等于0dB及以上对卫星信号进行精确检测(>95%),优于典型的目标检测方法.展开更多
针对大多数基于图神经网络(Graph Neural Network,GNN)的社区搜索方法中存在的时间开销巨大和“搭便车”效应问题,本文提出一种基于图组合优化的高效社区搜索模型(Efficient Community Search Based on Graph Combinatorial Optimizatio...针对大多数基于图神经网络(Graph Neural Network,GNN)的社区搜索方法中存在的时间开销巨大和“搭便车”效应问题,本文提出一种基于图组合优化的高效社区搜索模型(Efficient Community Search Based on Graph Combinatorial Optimization,CS-ROMF).该模型设计基于GNN的社区定位器来快速定位查询节点的潜在社区,减少时间开销.在此基础上设计基于强化学习(Reinforcement Learning,RL)的社区优化器调整候选社区的结构,减轻“搭便车”效应.在5个具有真实社区的数据集上进行大量实验,结果表明CS-ROMF在所有评估指标上均优于基线模型.其中,相比结果最好的基线模型,CS-ROMF在F_(1)值、Jaccard值以及NMI上分别最高提升14.99%、20.67%和21.37%,表明CS-ROMF减轻了“搭便车”效应.同时,CS-ROMF能够显著提升搜索效率,其运行速度比基于GNN的基线模型最多快10倍.展开更多
文摘To improve the accuracy of short text matching,a short text matching method with knowledge and structure enhancement for BERT(KS-BERT)was proposed in this study.This method first introduced external knowledge to the input text,and then sent the expanded text to both the context encoder BERT and the structure encoder GAT to capture the contextual relationship features and structural features of the input text.Finally,the match was determined based on the fusion result of the two features.Experiment results based on the public datasets BQ_corpus and LCQMC showed that KS-BERT outperforms advanced models such as ERNIE 2.0.This Study showed that knowledge enhancement and structure enhancement are two effective ways to improve BERT in short text matching.In BQ_corpus,ACC was improved by 0.2%and 0.3%,respectively,while in LCQMC,ACC was improved by 0.4%and 0.9%,respectively.
文摘复杂电磁环境下卫星信号往往淹没在背景和噪声中,传统的信号检测算法在没有准确先验知识的情况下性能急剧降低,目前基于深度学习的信号检测算法往往需要依赖专家经验的数据后处理步骤,无法对信号进行端到端检测.针对上述缺陷,提出一种基于DETR_S(DEtection with TRansformer on Signal)的卫星信号智能检测方法.DETR_S以编码器-解码器架构为基础,利用Transformer网络全局建模能力捕获频谱信息,采用多头自注意力机制有效改善频谱信息长距离依赖的问题.基于匈牙利算法的预测框匹配模块摒弃了非极大值抑制的数据后处理步骤,将信号检测问题转变为集合预测问题,使模型并行输出检测结果.引入信号重构模块,将频谱重构损失函数加入损失函数中,辅助模型挖掘频谱深层表征,提升信号检测性能.实验结果表明,在仅使用信号频谱幅度信息条件下,DETR_S能够在信噪比等于0dB及以上对卫星信号进行精确检测(>95%),优于典型的目标检测方法.
文摘针对大多数基于图神经网络(Graph Neural Network,GNN)的社区搜索方法中存在的时间开销巨大和“搭便车”效应问题,本文提出一种基于图组合优化的高效社区搜索模型(Efficient Community Search Based on Graph Combinatorial Optimization,CS-ROMF).该模型设计基于GNN的社区定位器来快速定位查询节点的潜在社区,减少时间开销.在此基础上设计基于强化学习(Reinforcement Learning,RL)的社区优化器调整候选社区的结构,减轻“搭便车”效应.在5个具有真实社区的数据集上进行大量实验,结果表明CS-ROMF在所有评估指标上均优于基线模型.其中,相比结果最好的基线模型,CS-ROMF在F_(1)值、Jaccard值以及NMI上分别最高提升14.99%、20.67%和21.37%,表明CS-ROMF减轻了“搭便车”效应.同时,CS-ROMF能够显著提升搜索效率,其运行速度比基于GNN的基线模型最多快10倍.
