基于自主学习与SCAD-Net正则化的回归模型

众多基因生物标志物选择方法常因研究样本较少而不能直接用于临床诊断.于是有学者提出整合不同基因表达数据同时保留生物信息完整性的方法.然而,由于存在批量效应,导致直接整合不同基因表达数据可能会增加新的系统误差.针对上述问题,提出一个融合自主学习与SCAD-Net正则化的分析框架.一方面,自主学习方法能够先从低噪声样本中学习出基础模型,然后再通过高噪声样本学习使得模型更加稳健,从而避免批量效应;另一方面,SCAD-Net正则化融合了基因表达数据与基因间的交互信息,可以实现更好的特征选择效果.不同情形下的模拟数据以及在乳腺癌细胞系数据集上的结果表明,基于自主学习与SCAD-Net正则化的回归模型在处理高维复杂网络数据集时具有更好的预测效果.     

Intelligent fault diagnosis for rolling bearings based on graph shift regularization with directed graphs

Graph shift regularization is a new and effective graph-based semi-supervised classification method, but its performance is closely related to the representation graphs. Since directed graphs can convey more information about the relationship between vertices than undirected graphs, an intelligent method called graph shift regularization with directed graphs (GSR-D) is presented for fault diagnosis of rolling bearings. For greatly improving the diagnosis performance of GSR-D, a directed and weighted k-nearest neighbor graph is first constructed by treating each sample (i.e., each vibration signal segment) as a vertex, in which the similarity between samples is measured by cosine distance instead of the commonly used Euclidean distance, and the edge weights are also defined by cosine distance instead of the commonly used heat kernel. Then, the labels of samples are considered as the graph signals indexed by the vertices of the representation graph. Finally, the states of unlabeled samples are predicted by finding a graph signal that has minimal total variation and satisfies the constraint given by labeled samples as much as possible. Experimental results indicate that GSR-D is better and more stable than the standard convolutional neural network and support vector machine in rolling bearing fault diagnosis, and GSR-D only has two tuning parameters with certain robustness.     

基于联合损失胶囊网络的换衣行人重识别

目前的行人重识别（Re-ID）研究主要集中在短时间情形,即一个人的衣着不太可能发生改变的情况。然而现实中更常见的是长时间的情况,这时一个人有很大的机会更换衣服,Re-ID模型应该考虑这种情况。为此,研究了一种基于联合损失胶囊网络的换衣行人重识别方法。所提方法基于换衣行人重识别胶囊网络ReIDCaps,使用与传统的标量神经元相比包含更多信息的矢量胶囊,用其长度表示行人身份信息,用其方向表示行人衣着信息;采用软嵌入注意力（SEA）防止模型过拟合;使用特征稀疏表示（FSR）机制提取具有判别性的特征;增加标签平滑正则化交叉熵损失与Circle Loss的联合损失以提高模型的泛化能力和鲁棒性。在三个换衣行人重识别数据集Celeb-reID、Celeb-reID-light和NKUP上进行实验,实验结果表明所提方法与目前已有的Re-ID方法相比具有一定优势。     

采用自适应字典学习的InSAR降噪方法

We consider the phase noise filtering problem for interferometric synthetic aperture radar (InSAR) based on the dictionary learning technique. Due to the non-convexity of the optimization problem is difficult to solve. By using the splitting technique and employing the augmented Lagrangian framework, we obtain a relaxed nonlinear constraint optimization problem with l₁-norm regularization which can be solved efficiently by the alternating direction method of multipliers (ADMM). Specifically, we firstly train dictionaries from the InSAR complex phase data, and then reconstruct the desired complex phase image from the sparse representation. Simulation results based on simulated and measured data show that this new InSAR phase noise reduction method has a much better performance than several classical phase filtering methods in terms of residual count, mean square error (MSE) and preservation of the fringe completeness.     

带约束的一维时频电磁联合反演研究

利用时频电磁资料可同时获取时间域及频率域电磁信息。相比单一时间域或频率域电磁信号,综合利用时频电磁信息可明显提高电性反演精度。为获得精细地电模型,基于正则化反演方法,融合数据约束、模型约束及步长约束,提出一种一维自适应时频电磁联合反演算法。该算法对时间域和频率域数据平衡因子及正则化因子采用自适应调整方案,同时将模型参数搜索范围约束与WOLFE准则相结合,在限制参数搜索范围的同时约束迭代步长,确保反演的稳定性和精度。使用经典层状模型、含噪声的页岩气地电模型及实测数据对该反演方法进行测试与分析,结果表明带约束的时频电磁联合反演方法可有效提高复杂模型电阻率反演的分辨率,对提高时频电磁方法油气藏识别精度具有实际意义。     

正则化稀疏模型

正则化稀疏模型在机器学习和图像处理等领域发挥着越来越重要的作用,它具有变量选择功能,可以解决建模中的过拟合等问题．Tibshirani 提出的 Lasso 使得正则化稀疏模型真正开始流行．文中总结了各种正则化稀疏模型,指出了各个稀疏模型被提出的原因、所具有的优点、适宜解决的问题及其模型的具体形式．最后,文中还指出了正则化稀疏模型未来的研究方向．     

基于多目标分层遗传模糊建模的磨矿过程溢流粒度软测量

提出一种基于多目标分层遗传算法的模糊系统对溢流粒度进行软测量, 该方法将模糊系统分为4 层, 即输入层、隶属度层、规则库层和系统集成层. 为了达到各层共同进化的目的, 设计遗传算法各层编码策略, 构建基于平均绝对百分误差和均方根误差的优化目标函数, 并采用该函数计算各层个体的适应度. 鉴于模糊模型训练过程中可能出现异常解, 将L-M 贝叶斯正则化方法融入训练过程. 对磨矿生产数据的仿真实验验证了所提出方法的有效性.

     

联合模板先验概率和稀疏表示的目标跟踪

目的 虽然基于稀疏表示的目标跟踪方法表现出了良好的跟踪效果,但仍然无法彻底解决噪声、旋转、遮挡、运动模糊、光照和姿态变化等复杂背景下的目标跟踪问题。针对遮挡、旋转、姿态变化和运动模糊问题,提出一种在粒子滤波框架内,基于稀疏表示和先验概率相结合的目标跟踪方法。方法 通过先验概率衡量目标模板的重要性,并将其引入到正则化模型中,作为模板更新的主要依据,从而获得一种新的候选目标稀疏表示模型。结果 在多个测试视频序列上,与多种流行算法相比,该算法可以达到更好的跟踪性能。在5个经典测试视频下的平均中心误差为6.77像素,平均跟踪成功率为97%,均优于其他算法。结论 实验结果表明,在各种含有遮挡、旋转、姿态变化和运动模糊的视频中,该算法可以稳定可靠地跟踪目标,适用于视频监控复杂场景下的目标跟踪。     

A cubic regularization algorithm for unconstrained optimization using line search and nonmonotone techniques

In recent years, cubic regularization algorithms for unconstrained optimization have been defined as alternatives to trust-region and line search schemes. These regularization techniques are based on the strategy of computing an (approximate) global minimizer of a cubic overestimator of the objective function. In this work we focus on the adaptive regularization algorithm using cubics (ARC) proposed in Cartis et al. [Adaptive cubic regularisation methods for unconstrained optimization. Part I: motivation, convergence and numerical results, Mathematical Programming A 127 (2011), pp. 245–295]. Our purpose is to design a modified version of ARC in order to improve the computational efficiency preserving global convergence properties. The basic idea is to suitably combine a Goldstein-type line search and a nonmonotone accepting criterion with the aim of advantageously exploiting the possible good descent properties of the trial step computed as (approximate) minimizer of the cubic model. Global convergence properties of the proposed nonmonotone ARC algorithm are proved. Numerical experiments are performed and the obtained results clearly show satisfactory performance of the new algorithm when compared to the basic ARC algorithm.