Steganalysis of neural networks based on parameter statistical bias

Many pretrained deep learning models have been released to help engineers and researchers develop deep learning-based systems or conduct research with minimall effort. Previous work has shown that at secret message can be embedded in neural network parameters without compromising the accuracy of the model. Malicious developers can, therefore, hide malware or other baneful information in pretrained models, causing harm to society. Hence, reliable detection of these vicious pretrained models is urgently needed. We analyze existing approaches for hiding messages and find that they will ineluctably cause biases in the parameter statistics. Therefore, we propose steganalysis methods for steganography on neural network parameters that extract statistics from benign and malicious models and build classifiers based on the extracted statistics. To the best of our knowledge, this is the first study on neural network steganalysis. The experimental results reveal that our proposed algorithm can effectively detect a model with an embedded message. Notably, our detection methods are still valid in cases where the payload of the stego model is low.     

基于作战环的不同节点攻击策略下的作战网络效能评估

基于作战环思想, 以作战网络抽象模型为基础, 围绕作战环的分类、定义、形式化描述、数学模型等基本问题开展了理论研究, 进一步丰富了作战环理论体系。同时, 以能否快速有效打击对方目标实体为根本依据, 建立了基于目标节点打击率和基于目标节点打击效率的作战网络效能评估指标。通过仿真实验, 将作战环相关理论运用于不同节点攻击策略下的作战网络效能评估, 为实际作战体系对抗中双方的攻击与防护提供应用指导, 也为进一步开展基于作战环的作战网络应用研究提供参考借鉴。     

基于码头闸口数据的集卡周转时间短时预测方法

为给车辆调度优化以及集卡预约系统设计提供重要参考,利用所采集的深圳市某港口的码头闸口数据,建立一种基于数据挖掘的集卡周转时间短时预测方法.首先,通过对码头闸口数据进行分析,获取车辆到达时间分布、任务类型、作业方式等集卡作业特征以及集卡在码头内的周转时间;在此基础上,利用循环神经网络并结合训练集数据,建立集卡作业特征与其周转时间之间的映射关系.其次,为减少随机波动对周转时间预测效果的影响,利用小波分解算法对循环神经网络拟合结果的残差进行高频噪声分离,并通过自回归模型拟合过滤后的低频序列.最后,将拟合后的循环神经网络与自回归模型进行结合,建立一种支持集卡周转时间短时预测的组合模型,并利用测试集数据进行有效性验证.结果 表明,相比单一的循环神经网络,该组合模型可以大幅提升预测精度.     

健身指导与实践校本课程开发程序研究

在国家师范类专业认证背景下,根据内江师范学院"三分式"人才培养改革模式,对健身指导与实践校本课程开发依据进行分析,对影响因素归纳总结,并对该课程开发程序进行研究.     

Predictive Event-Triggered Control for Disturbanced Wireless Networked Control Systems

The control and scheduling for wireless networked control system with packet dropout and disturbance are investigated. A prediction based event triggered control is proposed to reduce data transmissions while preserving the robustness against external disturbance. First, a trigger threshold is especially designed to maintain the difference of the estimated and actual states below a proper boundary when system suffers from packet dropout. Then a predictive controller is designed to compensate for packet dropouts by utilizing the packet-based control approach. The sufficient conditions to ensure the closed-loop system being uniformly ultimately bounded are derived, with consequently the controller gain method. Numerical examples illustrate the effectiveness of the proposed approach.     

目标电磁散射的面中心立方体网格FDTD方法

基于面中心立方体(face-centered cube, FCC)网格的空间结构, 由麦克斯韦方程出发, 推导了基于FCC网格的单轴各向异性介质完全匹配层(uniaxial anisotropic media perfectly matched layer, UPML)吸收边界条件以及近-远场外推边界条件的三维迭代式。通过典型算例, 先后验证了基于面中心立方体网格的时域有限差分(FCC-finite difference time domain, FCC-FDTD)方法的UPML吸收边界条件和近远场外推边界条件的正确性。最后通过计算金属球的后向雷达散射截面(radar cross section, RCS)比较了FCC-FDTD方法与传统FDTD方法的计算精度, 结果显示FCC-FDTD方法具有更高的计算精度。     

不同边界条件薄壁截锥壳的高阶振动特性研究

采用传递矩阵法研究了不同边界条件下薄壁截锥壳的高阶振动特性.基于Love壳体理论建立薄壁截锥壳振动微分方程,根据薄壁截锥壳子段间的状态向量,通过传递矩阵法得出整体传递矩阵,并用高精度的精细积分法计算固有频率,通过文献和有限元法进行验证,并分析了薄壁截锥壳在不同边界条件下的高阶振动特性.结果表明,不同边界条件下,采用传递矩阵法计算高阶固有频率与有限元法的计算结果基本一致.当轴向半波数增加时,频率明显增大;随着周向波数的增加,频率先减小后增大.固支-固支和简支-简支边界下在m=1和n=7处得到最小频率,固支-自由边界下在m=1和n=6处得到最小频率,三种边界下最小频率值分别为400.1、325.6和226.1 Hz;边界条件约束越多,最低阶固有频率越大.     

基于差异信息量的多源数据融合方法

D-S证据理论可应用于多源数据融合领域，但在处理高度冲突的证据时，可能会出现反直觉的结果.为解决这一问题，本文提出了差异信息量的概念及融合方法.首先，通过信息熵表明证据的相对重要性，采用散度获取证据可信度.然后利用证据可信度优化证据差异度以得到差异信息量，经过计算获取数据的最终权重，并将其作为D-S证据理论中的基本概率分配进行决策.在处理冲突证据、一致证据及不同数量证据等方面的数据融合问题时与其他方法对比，所提方法收敛更快，准确度更高.故障诊断的应用实例表明，所提方法的不确定性更小，优于现存的其他方法.     

Multifunctional artificial nacre via biomimetic matrix-directed mineralization

Natural nacre, one of the most studied biological structural materials with delicate hierarchical structures and extraordinary performance, has inspired the design and fabrication of artificial structural ceramics with high fracture toughness. However, to meet the diverse requirements of different applications, future structural materials must be multifunctional with superior mechanical properties, such as strength, hardness, and toughness. Herein, based on the matrix-directed mineralization method for producing biomimetic structural materials, we introduce nanoparticles with different inherent functions into the platelets of artificial nacre via the co-mineralization of aragonite and the nanoparticles. Besides their enhanced mechanical properties, the obtained artificial nacre materials also exhibit different functions depending on the type of the nanoparticles. To extend the versatility of this strategy, the effects of nanoparticles of different sizes and zeta potentials on mineralization are also analyzed. This universal strategy can be applied to the fabrication of other types of functionalized biomimetic structural ceramics that have potential applications in various fields, such as biomedical science.     

Evolutionary game analysis of low-carbon behavior credit supervision of logistics enterprises

Based on the low-carbon obligation fulfillment of Chinese logistics enterprises, this study constructs a tripartite evolutionary game model to analyze the evolutionary process of the interaction between the local government, logistics enterprises and the public in the process of low-carbon behavior credit supervision. Then using Netlogo software, a parameter simulation experiment is conducted to determine the optimal policy for improving the effect of supervision. The results are as follows: ① The combined influence of the local government and the public can effectively change the strategic choice of enterprises and promote the low-carbon behavior of enterprises. ② In terms of improving the effect of supervision, reducing the cost of government supervision would have a highly significant effect, and reducing the cost of the public would be counterproductive. ③ Increasing the government subsidies to enterprises and the government’s fines to enterprises both have a significant effect, and the effect of improving the former is better. However, increasing the severity of higher-level governments punishing local governments will reduce the stability of the system. ④ Supervision can be more effective by increasing the public’s impact on enterprises’ earnings rather than by increasing government subsidies to the public.