Privacy-preserving federated learning framework with dynamic weight aggregation by Zuobin YING, Yichen FANG, Yiwen ZHANG

“…There are two problems with the privacy-preserving federal learning framework under an unreliable central server.① A fixed weight, typically the size of each participant’s dataset, is used when aggregating distributed learning models on the central server.However, different participants have non-independent and homogeneously distributed data, then setting fixed aggregation weights would prevent the global model from achieving optimal utility.② Existing frameworks are built on the assumption that the central server is honest, and do not consider the problem of data privacy leakage of participants due to the untrustworthiness of the central server.To address the above issues, based on the popular DP-FedAvg algorithm, a privacy-preserving federated learning DP-DFL algorithm for dynamic weight aggregation under a non-trusted central server was proposed which set a dynamic model aggregation weight.The proposed algorithm learned the model aggregation weight in federated learning directly from the data of different participants, and thus it is applicable to non-independent homogeneously distributed data environment.In addition, the privacy of model parameters was protected using noise in the local model privacy protection phase, which satisfied the untrustworthy central server setting and thus reduced the risk of privacy leakage in the upload of model parameters from local participants.Experiments on dataset CIFAR-10 demonstrate that the DP-DFL algorithm not only provides local privacy guarantees, but also achieves higher accuracy rates with an average accuracy improvement of 2.09% compared to the DP-FedAvg algorithm models.…”
Get full text

Federated learning resource management for energy-constrained industrial IoT devices by Shaoshuai FAN, Jianbo WU, Hui TIAN

“…Given the impact of limited wireless resources, a dynamic multi-dimensional resource joint management algorithm was proposed, which intended to tackle the problem of device failure and training interruption caused by the limited battery energy in federated learning network in industrial Internet of things (IIoT).Firstly, the optimization problem was decoupled into battery energy allocation, equipment resource allocation and communication resource allocation sub-problems which were interdependent with the goal of maximizing the fixed-time learning accuracy.Then, the equipment transmission and computing resource allocation problem were solved based on particle swarm optimization algorithm under the given energy budget.Thereafter, the resource block iterative matching algorithm was proposed to optimize the optimal communication resource allocation strategy.Finally, the online energy allocation algorithm was proposed to adjust the energy budget allocation.Simulation results validate the proposed algorithm can improve the model learning accuracy compared with other benchmarks, and can perform better in energy shortage scenarios.…”
Get full text

A Bicriteria Approach Identifying Nondominated Portfolios by Javier Pereira, Broderick Crawford, Fernando Paredes, Ricardo Soto

“…In order to improve the former algorithm, a process finding an optimal solution of a monocriteria version of this problem is proposed, which is further used as a first feasible solution aiding to find nondominated solutions more rapidly. …”
Get full text

Adaptive Neural Network Iterative Sliding Mode Course Tracking Control for Unmanned Surface Vessels by Chunbo Zhao, Huaran Yan, Deyi Gao, Renqiang Wang, Qinrong Li

“…Therefore, the fusion of multiple algorithms can be applied to improve the performance of the control system.…”
Get full text

Study on the Speed Ratio Control of Fuzzy PID Control for EM- CVT based on PSO by Zhang Lanchun, Liu Sen, Wang Wei, Shen Hongsheng, Wang Shanfeng

“…The Fuzzy PID speed ratio controller model is set up,whose parameter factors are optimized by the PSO algorithm to realize the optimization of the speed ratio control system. …”
Get full text

DMSO-free cryopreservation of hiPSC-derived cardiomyocytes: low temperature characterization and protocol development by Akshat S. Mallya, Tessa Burrows, Jeanne Hsieh, Troy Louwagie, James R. Dutton, Brenda M. Ogle, Allison Hubel

“…A differential evolution (DE) algorithm was utilized to determine the optimal composition of a mixture of a sugar, sugar alcohol and amino acid to replace DMSO as the CPA. …”
Get full text

Safety helmet detection methods in heavy machinery factory by Liu Baoju, Wei Xiangqian, Chen Qingshan, Liu Jiaqi, Chen Ye, Yu Peng, Lei shi, Hu Yongfeng

“…When compared with mainstream object recognition algorithms such as SSD, Faster RCNN, and various YOLO versions, the optimized model shows its superiority. …”
Get full text

UHVDC Transmission Line Fault Identification Method Based on Generalized Regression Neural Network by XIE Jia, LIU Feng, KE Yanguo, YIN Zhen, RUAN Wei, YAO Jinming

“…Secondly, the chaos quantum particle swarm optimization (CQPSO) algorithm is used to optimize the parameters of the generalized regression neural network, form an ideal network model based on the principle of the lowest fitness function, and better learn the fault characteristics of the ultra-high voltage DC transmission line. …”
Get full text

Simulation of incremental update of electronic document information based on big data technology by Zhiyuan Jin, Qi Zhang, Tiejun Pan

“…The network structure optimization is mainly achieved by increasing the number of neurons in the hidden layer, and the BP algorithm and gradient descent method are used to iteratively calculate until the model converges, so as to improve the model’s ability to capture the dynamic features of incremental data. …”
Get full text

LIGHTWEIGHT OF THE VEHICLE B PILLAR BASED ON SIX SIGMA ROBUST DESIGN UNDER SIDE IMPACT by QIU RuiBin, CHEN Yuan, LEI Fei, YANG ShaoYong, LI YongJun

“…In this paper,by considering uncertainty factors,taking side impact of a car as an example,first of all,an accurate finite element model is established and verified according to relevant crash regulations,then,after building an approximate model and a deterministic optimization model,the optimal solution is obtained by sequential quadratic programming algorithm and the reliability analysis is used,finally the original vehicle B pillar is accomplished lightweight research and improvement by using six sigma robust design method. …”
Get full text

Deep Reinforcement Learning for UAV Intelligent Mission Planning by Longfei Yue, Rennong Yang, Ying Zhang, Lixin Yu, Zhuangzhuang Wang

“…Then, the SEAD intelligent planning model based on the proximal policy optimization (PPO) algorithm is established and a general intelligent planning architecture is proposed. …”
Get full text

Research on collaborative scheduling strategies of multi-agent agricultural machinery groups by Ziyi Wang, Fan Zhang, Shiji Ma, Hailong Wang, Shunyao Zhang, Xiaozhong Gao

“…Subsequently, the network model is trained to improve its solution capabilities using the policy gradient algorithm (REINFORCE). …”
Get full text

Vehicle Routing Problem for Collaborative Multidepot Petrol Replenishment under Emergency Conditions by Guangcan Xu, Qiguang Lyu

“…As a method to solve the model, genetic variation of multiobjective particle swarm optimization algorithm is considered. …”
Get full text

基于进化计算的多任务可调机构刚体导引综合 by 杨志刚, 徐礼钜, 林光春, 张均富

“…The method of rigid-body guidance synthesis of adjustable mechanisms for multiple alternation tasks based on evolutionary computation is proposed.The mechanism optimization model of multi-tasks is established,and the global optimization solution will be gained easily by using evolutionary computation.The algorithm of evolutionary computation is improved.A method that can adaptively adjust mutation rate and mutation value according to fitness of individual is proposed,which can effectively improve the evolutionary speed and the accuracy of the solutions.Two numeric examples are given to illustrate the method.…”
Get full text

考虑润滑状态的摆动活齿传动多目标优化设计 by 邓志平, 张均富

“…The global optimal solution is obtained based on chaotic particle swarm optimization algorithm. …”
Get full text

An integrated approach of feature selection and machine learning for early detection of breast cancer by Jing Zhu, Zhenhang Zhao, Bangzheng Yin, Canpeng Wu, Chan Yin, Rong Chen, Youde Ding

“…Optimizing hyperparameters of five models using the Particle Swarm Optimization (PSO) algorithm. …”
Get full text

Research on the application of deep learning based video recognition for power plant leakage and dripping by Zhang Yuan, Si Yuan

“…Then, by combining semantic segmentation, data augmentation, attention mechanisms, and changing activation functions with convolutional neural networks, the YOLOv5 algorithm is deeply optimized, including improvements in training strategies and model evaluation adjustments. …”
Get full text

Research on Short-Term Load Forecasting of LSTM Regional Power Grid Based on Multi-Source Parameter Coupling by Bo Li, Yaohua Liao, Siyang Liu, Chao Liu, Zhensheng Wu

“…In order to further optimize the performance of the LSTM model, the IPSO algorithm, and linear difference decreasing inertia weight are introduced to improve the global optimization ability and convergence speed of the PSO algorithm and reduce the risk of local optimal solutions. …”
Get full text

基于遗传-二次规划法的封闭差动人字齿轮传动系统混合优化设计 by 朱增宝, 朱如鹏, 尹敏

“…Using the positive transmission way of modified gear,a mathematical optimization model of the encased differential herringbone gear transmission system is set up by taking the lightest gear weight of this transmission system as the optimization objective.Taking the advantage of its optimization function of genetic algorithm and quadratic programming algorithm in MATLAB toolbox,the compound optimization design is carried out.The research results show that the problem of which the initial value of quadratic programming is difficult to determine is solved well by using the results of the genetic algorithm as the initial value of quadratic programming optimization.Compared with the genetic algorithm or quadratic programming,the optimization design variables of this compound optimization are obviously improved and the gear weight of the system is markedly reduced.The compound optimization method provides a simple and practical solution to obtain precise global optimal solution of the large-scale nonlinear constraint optimization.…”
Get full text

Design of the College Students’ Music Big Data Management System Based on Computer Assistance by Peng Cheng

“…With the rapid development of science and technology, the college music big data management system also needs a computer-aided data model for optimization. In order to improve the efficiency of computer-aided teaching of the music big data management system in colleges and universities, this paper analyzes and processes the music database of music software based on the computer-aided Bayesian algorithm and establishes a Bayesian model. …”
Get full text