Distributed Nash Equilibrium Seeking for General Networked Games With Bounded Disturbances

This paper is concerned with anti-disturbance Nash equilibrium seeking for games with partial information. First, reduced-order disturbance observer-based algorithms are proposed to achieve Nash equilibrium seeking for games with first-order and second-order players, respectively. In the developed algorithms, the observed disturbance values are included in control signals to eliminate the influence of disturbances, based on which a gradient-like optimization method is implemented for each player. Second, a signum function based distributed algorithm is proposed to attenuate disturbances for games with second-order integrator-type players. To be more specific, a signum function is involved in the proposed seeking strategy to dominate disturbances, based on which the feedback of the velocity-like states and the gradients of the functions associated with players achieves stabilization of system dynamics and optimization of players’ objective functions. Through Lyapunov stability analysis, it is proven that the players’ actions can approach a small region around the Nash equilibrium by utilizing disturbance observer-based strategies with appropriate control gains. Moreover, exponential (asymptotic) convergence can be achieved when the signum function based control strategy (with an adaptive control gain) is employed. The performance of the proposed algorithms is tested by utilizing an integrated simulation platform of virtual robot experimentation platform (V-REP) and MATLAB.      

以双目图像互为监督的相容解迭代优选算法

在计算机视觉领域的双目立体匹配方向, 基于神经网络的深度学习算法需要场景数据集进行训练, 泛化能力差. 针对这两个问题, 根据神经网络能够模拟函数的特点, 提出一种无需在数据集上训练, 以双目图像互为监督的深度场景相容解迭代优选算法. 该算法使用场景位置猜测网络模拟关于当前双目图像的深度场景相容位置空间, 用与该网络匹配的互监督损失函数通过梯度下降法指导该网络在输入双目图像上迭代学习, 搜索深度场景相容位置空间中的可行解, 整个算法过程无需在数据集上训练. 与CREStereo、PCW-Net、CFNet等算法在Middlebury标准数据集图像上的对比实验表明, 该算法在非遮挡区域的平均误匹配率为2.52%, 在所有区域的平均误匹配率为7.26%, 比对比实验中的其他算法有更低的平均误匹配率.     

一种结合机器学习的移动边缘计算的切换预测方法

随着移动通信的发展,减少通信延时成为关键性问题,因此,提出了一种使用机器学习方法的移动边缘计算(MEC)移动性管理。移动性决策基于参考信号接收功率(RSRP)值和不确定性预测器。使用神经网络建立预测器,输出不同相邻单元的RSRP平均值和标准偏差,推导了切换概率的封闭表达式。基于这些可能性,MEC服务器能够提前缓存用户服务,以便将切换期间的中断降至最低。实验结果表明,提出的方法能够满足实际需求。     

融合CNN-BiLSTM-Attention的集成学习价格预测

价格预测对于大宗农产品市场的稳定具有重要意义,但是大宗农产品价格与多种因素有着复杂的相关关系.针对当前价格预测中对数据完整性依赖性强与单一模型难以全面利用多种数据特征等问题,提出了一种将基于注意力机制的卷积双向长短期记忆神经网络(CNN-BiLSTM-Attention)、支持向量机回归(SVR)与LightGBM组合的增强式集成学习方法,并分别在包含历史交易、天气、汇率、油价等多种特征数据的数据集上进行了实验.实验以小麦和棉花价格预测为目标任务,使用互信息法进行特征选择,选择误差较低的CNN-BiLSTM-Attention模型作为基模型,与机器学习模型通过线性回归进行增强式集成学习.实验结果表明该集成学习方法在小麦及棉花数据集上预测结果的均方根误差(RMSE)值分别为12.812, 74.365,较之3个基模型分别降低11.00%, 0.94%、4.44%,1.99%与13.03%, 4.39%,能够有效降低价格预测的误差.     

基于预测对抗网络的图像二分类模型

正未标记学习仅使用无标签样本和正样本训练一个二分类器, 而生成式对抗网络(generative adversarial networks, GAN)中通过对抗性训练得到一个图像生成器. 为将GAN的对抗训练方法迁移到正未标记学习中以提升正未标记学习的效果, 可将GAN中的生成器替换为分类器C, 在无标签数据集中挑选样本以欺骗判别器D, 对C与D进行迭代优化. 本文提出基于以Jensen-Shannon散度(JS散度)为目标函数的JS-PAN模型. 最后, 结合数据分布特点及现状需求, 说明了PAN模型在医疗诊断图像二分类应用的合理性及高性能. 在MNIST, CIFAR-10数据集上的实验结果显示: KL-PAN模型与同类正未标记学习模型对比有更高的精确度(ACC)及F1-score; 对称化改进后, JS-PAN模型在两个指标上均有所提升, 因此JS-PAN模型的提出更具有合理性. 在Med-MNIST的3个子图像数据集上的实验显示: KL-PAN模型与4个benchmark有监督模型有几乎相同的ACC, JS-PAN也有更高表现. 因此, 综合PAN模型的出色分类效果及医疗诊断数据的分布特征, PAN作为半监督学习方法可获得更快、更好的效果, 在医学图像的二分类的任务上具有更高的性能.     

Almost sure stability of stochastic neutral Cohen–Grossberg neural networks with Lévy noise and time-varying delays

The almost sure stability for the stochastic neutral Cohen–Grossberg neural networks (SNCGNNs) with Lévy noise, time-varying delays, and Markovian switching would be deliberated in this article. By means of the nonnegative semimartingale convergence theorem (NSCT), the neutral Itô formula, M-matrix method, and selecting appropriate Lyapunov function, several almost sure stability criterions for the SNCGNNs could be derived. Moreover, according to the M-matrix theory, the upper bounds of the coefficients at any mode are given. Finally, two examples and numerical simulations verify the correctness of theoretical analysis for the stability criterions proposed in the article.     

Ultrawideband coupled relative positioning algorithm applicable to flight controller for multidrone collaboration

In this study, we introduce a loosely coupled relative position estimation method that utilizes a decentralized ultrawideband (UWB), Global Navigation Support System and inertial navigation system for flight controllers (FCs). Key obstacles to multidrone collaboration include relative position errors and the absence of communication devices. To address this, we provide an extended Kalman filter-based algorithm and module that correct distance errors by fusing UWB data acquired through random communications. Via simulations, we confirm the feasibility of the algorithm and verify its distance error correction performance according to the amount of communications. Real-world tests confirm the algorithm's effectiveness on FCs and the potential for multidrone collaboration in real environments. This method can be used to correct relative multidrone positions during collaborative transportation and simultaneous localization and mapping applications.     

不对称约束多人非零和博弈的自适应评判控制

本文针对连续时间非线性系统的不对称约束多人非零和博弈问题, 建立了一种基于神经网络的自适应评判控制方法. 首先, 本文提出了一种新颖的非二次型函数来处理不对称约束问题, 并且推导出最优控制律和耦合Hamilton-Jacobi方程. 值得注意的是, 当系统状态为零时, 最优控制策略是不为零的, 这与以往不同. 然后, 通过构建单一评判网络来近似每个玩家的最优代价函数, 从而获得相关的近似最优控制策略. 同时, 在评判学习期间发展了一种新的权值更新规则. 此外, 通过利用Lyapunov理论证明了评判网络权值近似误差和闭环系统状态的稳定性. 最后, 仿真结果验证了本文所提方法的有效性     

异质信息网络高阶层次化嵌入学习与推荐预测

异质信息网络是一种异质数据表示形式,如何融合异质数据复杂语义信息,是推荐系统面临的挑战之一.利用弱关系具有的丰富语义和信息传递能力,构建一种面向推荐系统的异质信息网络高阶嵌入学习框架,主要包括:初始化信息嵌入、高阶信息嵌入聚合与推荐预测3个模块.初始化信息嵌入模块首先采用基于弱关系的异质信息网络最佳信任路径筛选算法,有效地避免在全关系异质信息网络中,采样固定数量邻居造成的信息损失,其次利用新定义的基于多头图注意力的多任务共享特征重要性度量因子,筛选出节点的语义信息,并结合交互结构,有效地表征网络节点;高阶信息嵌入聚合模块通过融入弱关系及网络嵌入对知识良好的表征能力,实现高阶信息表达,并利用异质信息网络的层级传播机制,将被采样节点的特征聚合到待预测节点;推荐预测模块利用高阶信息的影响力推荐方法,实现了推荐任务.该框架具有嵌入节点类型丰富、融合共享属性和隐式交互信息等特点.最后,实验验证UI-HEHo学习框架可有效地改善评级预测的准确性,以及推荐生成的针对性、新颖性和多样性,尤其是在数据稀疏的应用场景中,具有良好的推荐效果.     

Rumor detection with self-supervised learning on texts and social graph

Rumor detection has become an emerging and active research field in recent years. At the core is to model the rumor characteristics inherent in rich information, such as propagation patterns in social network and semantic patterns in post content, and differentiate them from the truth. However, existing works on rumor detection fall short in modeling heterogeneous information, either using one single information source only (e.g., social network, or post content) or ignoring the relations among multiple sources (e.g., fusing social and content features via simple concatenation).Therefore, they possibly have drawbacks in comprehensively understanding the rumors, and detecting them accurately. In this work, we explore contrastive self-supervised learning on heterogeneous information sources, so as to reveal their relations and characterize rumors better. Technically, we supplement the main supervised task of detection with an auxiliary self-supervised task, which enriches post representations via post self-discrimination.Specifically, given two heterogeneous views of a post (i.e., representations encoding social patterns and semantic patterns), the discrimination is done by maximizing the mutual information between different views of the same post compared to that of other posts. We devise cluster-wise and instance-wise approaches to generate the views and conduct the discrimination, considering different relations of information sources. We term this framework as self-supervised rumor detection (SRD). Extensive experiments on three real-world datasets validate the effectiveness of SRD for automatic rumor detection on social media.