基于深度限制波尔兹曼机的辐射源信号识别

针对电子侦察中使用常规参数难以有效识别复杂体制雷达信号的问题,提出利用深度限制波尔兹曼机对辐射源识别的模型。模型由多个限制波尔兹曼机组成,通过逐层自底向上无监督学习获得初始参数,并用后向传播算法对整个模型进行有监督的参数微调,利用Softmax进行分类识别。通过仿真实验表明该模型能对辐射源进行有效的特征提取和分类识别,具有较高的识别精度和较强的鲁棒性。     

结合多视角学习与一致性表征的人脸伪造检测

现有的人脸伪造检测方法通常在已知域上表现较好,但面临过拟合的风险,在应对未知场景时无法保持良好的检测能力。为解决此问题,提出一种结合多视角学习与一致性表征的人脸伪造检测框架。为捕获更全面的伪造痕迹,将输入图像转换为两种互补视角并采用双流骨干网络进行多视角特征学习。引入一致性度量,以补丁级监督的方式明确约束不同视角输出的局部特征的相似度。为提高检测精度,采用特征分解策略进一步优化伪造特征,减少不相关因素的干扰,并以伪造特征空间的决策作为最终的预测结果。在基准数据集上进行的大量实验表明,所提出的方法优于现有的主流模型,具备良好的跨域泛化能力。     

矩阵变换特征与码序列联合学习的卷积码识别方法

现有基于深度学习的卷积码识别方法仍存在参数规模较大、识别性能较弱等不足。针对该问题,提出了一种基于矩阵变换特征与码序列联合学习的卷积码识别方法。将接收到的码字序列排列成矩阵形式,利用软信息剔除可靠性较低的码字,通过一种新的矩阵变换算法得到特征矩阵。在识别时,将原始码字矩阵和特征矩阵输入到具有多模态数据联合学习能力的网络模型,在神经网络中完成特征的提取融合与卷积码的识别。仿真结果表明,所提方法性能明显优于现有基于深度学习的识别方法,特别是对于高码率卷积码;当码率较低时,同样优于传统识别方法。当信噪比达到5 dB时,25种不同参数卷积码的识别率均可达到100%。     

基于神经网络的超精密机床伺服进给非线性模型辨识

以双轴Ｔ型结构的超精密金刚石车床的伺服进给系统为研究对象,采用自构造神经网络技术,建立了系统的非线性动态数学模型,为系统非线性控制与补偿提供参考模型     

遥测参数数据载荷状态判别集成学习方法

针对载荷单机设备遥测参数维度高、数据量大、存在类别不平衡、无法直观判别单机设备运行情况等问题,考虑到航天任务对可解释性的要求,提出一种基于信息增益参数特征选择和集成学习方法的载荷单机状态快速识别方法。采用统计量性质和信息增益子集搜索方法对遥测数据进行特征筛选降维,通过集成学习模型算法实现载荷单机设备状态的自适应识别分类。所提方法将信息增益的参数分类信息量评价准则和集成学习拟合能力强、类别不平衡下准确率高和抗噪能力强等优点相结合,兼顾模型特征和结果的可解释性,提供了重点参数发现功能。采用科学卫星任务真实载荷遥测参数数据对该方法进行了验证,整体识别准确率高于90%,少数样本亦可准确识别,整体效果可达到在轨任务要求,证明了所提方法的有效性和实用性。     

无监督特征学习的人体活动识别

针对人的局限性可能会导致在提取特征中丢失重要信息,从而影响最终的识别效果问题,提出无监督特征学习技术的惯性传感器特征提取方法。其核心思想是使用无监督特征学习方法学习多个特征映射,再将所有特征映射拼接起来形成最终的特征计算方法。其优点是不会造成重要信息的损失,而且可以显著减少所使用的无监督特征学习模型的规模。为了验证所提出的特征提取方法在活动识别中的有效性,运用一个公开的活动识别数据集,使用三种常用无监督模型进行特征提取,并使用支持向量机进行活动识别。实验结果表明,特征提取方法取得了良好的效果,与其他方法相比具有一定的优势。     

A misclassification cost‐minimizing evolutionary–neural classification approach

Machine learning algorithms that incorporate misclassification costs have recently received considerable attention. In this paper, we use the principles of evolution to develop and test an evolutionary/genetic algorithm (GA)‐based neural approach that incorporates asymmetric Type I and Type II error costs. Using simulated, real‐world medical and financial data sets, we compare the results of the proposed approach with other statistical, mathematical, and machine learning approaches, which include statistical linear discriminant analysis, back‐propagation artificial neural network, integrated cost preference‐based linear mathematical programming‐based minimize squared deviations, linear integrated cost preference‐based GA, decision trees (C 5.0, and CART), and inexpensive classification with expensive tests algorithm. Our results indicate that the proposed approach incorporating asymmetric error costs results in equal or lower holdout sample misclassification cost when compared with the other statistical, mathematical, and machine learning misclassification cost‐minimizing approaches. © 2006 Wiley Periodicals, Inc. Naval Research Logistics, 2006.     

学员英语学习动机激发之管见

学习动机是直接推动学生进行学习的一种内在动力。学习者的动机和态度作为非智力因素对于英语学习起着相当大的作用,在对军校学员英语学习动机深入了解的基础上,提出了激发学员英语学习动机的具体对策。     

Allocation of quality improvement targets based on investments in learning

Purchased materials often account for more than 50% of a manufacturer's product nonconformance cost. A common strategy for reducing such costs is to allocate periodic quality improvement targets to suppliers of such materials. Improvement target allocations are often accomplished via ad hoc methods such as prescribing a fixed, across‐the‐board percentage improvement for all suppliers, which, however, may not be the most effective or efficient approach for allocating improvement targets. We propose a formal modeling and optimization approach for assessing quality improvement targets for suppliers, based on process variance reduction. In our models, a manufacturer has multiple product performance measures that are linear functions of a common set of design variables (factors), each of which is an output from an independent supplier's process. We assume that a manufacturer's quality improvement is a result of reductions in supplier process variances, obtained through learning and experience, which require appropriate investments by both the manufacturer and suppliers. Three learning investment (cost) models for achieving a given learning rate are used to determine the allocations that minimize expected costs for both the supplier and manufacturer and to assess the sensitivity of investment in learning on the allocation of quality improvement targets. Solutions for determining optimal learning rates, and concomitant quality improvement targets are derived for each learning investment function. We also account for the risk that a supplier may not achieve a targeted learning rate for quality improvements. An extensive computational study is conducted to investigate the differences between optimal variance allocations and a fixed percentage allocation. These differences are examined with respect to (i) variance improvement targets and (ii) total expected cost. For certain types of learning investment models, the results suggest that orders of magnitude differences in variance allocations and expected total costs occur between optimal allocations and those arrived at via the commonly used rule of fixed percentage allocations. However, for learning investments characterized by a quadratic function, there is surprisingly close agreement with an “across‐the‐board” allocation of 20% quality improvement targets. © John Wiley & Sons, Inc. Naval Research Logistics 48: 684–709, 2001     

一种改进的BP算法

ＢＰ算法是目前应用极为广泛的神经网络算法,但它也存在一些不足。文中提出了采用共轭梯度法及黄金分割相结合的改进ＢＰ算法（ＭＢＰ）,自适应调整学习率,提高了运算速度,减少了迭代次数。最后将标准ＢＰ算法和ＭＢＰ算法进行了比较,仿其结果表明：ＭＢＰ算法的学习次数和收敛速度得到极大改善。