WebFLex:A Framework for Web Browsers-Based Peer-to-Peer Federated Learning Systems Using WebRTC

Scalability and information personal privacy are vital for training and deploying large-scale deep learning models.Federated learning trains models on exclusive information by aggregating weights from various devices and taking advantage of the device-agnostic environment of web browsers.Nevertheless,relying on a main central server for internet browser-based federated systems can prohibit scalability and interfere with the training process as a result of growing client numbers.Additionally,information relating to the training dataset can possibly be extracted from the distributed weights,potentially reducing the privacy of the local data used for training.In this research paper,we aim to investigate the challenges of scalability and data privacy to increase the efficiency of distributed training models.As a result,we propose a web-federated learning exchange(WebFLex)framework,which intends to improve the decentralization of the federated learning process.WebFLex is additionally developed to secure distributed and scalable federated learning systems that operate in web browsers across heterogeneous devices.Furthermore,WebFLex utilizes peer-to-peer interactions and secure weight exchanges utilizing browser-to-browser web real-time communication(WebRTC),efficiently preventing the need for a main central server.WebFLex has actually been measured in various setups using the MNIST dataset.Experimental results show WebFLex’s ability to improve the scalability of federated learning systems,allowing a smooth increase in the number of participating devices without central data aggregation.In addition,WebFLex can maintain a durable federated learning procedure even when faced with device disconnections and network variability.Additionally,it improves data privacy by utilizing artificial noise,which accomplishes an appropriate balance between accuracy and privacy preservation.

Artificial Intelligence Meets Flexible Sensors:Emerging Smart Flexible Sensing Systems Driven by Machine Learning and Artificial Synapses

The recent wave of the artificial intelligence(AI)revolution has aroused unprecedented interest in the intelligentialize of human society.As an essential component that bridges the physical world and digital signals,flexible sensors are evolving from a single sensing element to a smarter system,which is capable of highly efficient acquisition,analysis,and even perception of vast,multifaceted data.While challenging from a manual perspective,the development of intelligent flexible sensing has been remarkably facilitated owing to the rapid advances of brain-inspired AI innovations from both the algorithm(machine learning)and the framework(artificial synapses)level.This review presents the recent progress of the emerging AI-driven,intelligent flexible sensing systems.The basic concept of machine learning and artificial synapses are introduced.The new enabling features induced by the fusion of AI and flexible sensing are comprehensively reviewed,which significantly advances the applications such as flexible sensory systems,soft/humanoid robotics,and human activity monitoring.As two of the most profound innovations in the twenty-first century,the deep incorporation of flexible sensing and AI technology holds tremendous potential for creating a smarter world for human beings.

FL-EASGD:Federated Learning Privacy Security Method Based on Homomorphic Encryption

Federated learning ensures data privacy and security by sharing models among multiple computing nodes instead of plaintext data.However,there is still a potential risk of privacy leakage,for example,attackers can obtain the original data through model inference attacks.Therefore,safeguarding the privacy of model parameters becomes crucial.One proposed solution involves incorporating homomorphic encryption algorithms into the federated learning process.However,the existing federated learning privacy protection scheme based on homomorphic encryption will greatly reduce the efficiency and robustness when there are performance differences between parties or abnormal nodes.To solve the above problems,this paper proposes a privacy protection scheme named Federated Learning-Elastic Averaging Stochastic Gradient Descent(FL-EASGD)based on a fully homomorphic encryption algorithm.First,this paper introduces the homomorphic encryption algorithm into the FL-EASGD scheme to preventmodel plaintext leakage and realize privacy security in the process ofmodel aggregation.Second,this paper designs a robust model aggregation algorithm by adding time variables and constraint coefficients,which ensures the accuracy of model prediction while solving performance differences such as computation speed and node anomalies such as downtime of each participant.In addition,the scheme in this paper preserves the independent exploration of the local model by the nodes of each party,making the model more applicable to the local data distribution.Finally,experimental analysis shows that when there are abnormalities in the participants,the efficiency and accuracy of the whole protocol are not significantly affected.

A Deep Learning Approach to Shape Optimization Problems for Flexoelectric Materials Using the Isogeometric Finite Element Method

A new approach for flexoelectricmaterial shape optimization is proposed in this study.In this work,a proxymodel based on artificial neural network(ANN)is used to solve the parameter optimization and shape optimization problems.To improve the fitting ability of the neural network,we use the idea of pre-training to determine the structure of the neural network and combine different optimizers for training.The isogeometric analysis-finite element method(IGA-FEM)is used to discretize the flexural theoretical formulas and obtain samples,which helps ANN to build a proxy model from the model shape to the target value.The effectiveness of the proposed method is verified through two numerical examples of parameter optimization and one numerical example of shape optimization.

Deep Reinforcement Learning-Based Task Offloading and Service Migrating Policies in Service Caching-Assisted Mobile Edge Computing

Emerging mobile edge computing(MEC)is considered a feasible solution for offloading the computation-intensive request tasks generated from mobile wireless equipment(MWE)with limited computational resources and energy.Due to the homogeneity of request tasks from one MWE during a longterm time period,it is vital to predeploy the particular service cachings required by the request tasks at the MEC server.In this paper,we model a service caching-assisted MEC framework that takes into account the constraint on the number of service cachings hosted by each edge server and the migration of request tasks from the current edge server to another edge server with service caching required by tasks.Furthermore,we propose a multiagent deep reinforcement learning-based computation offloading and task migrating decision-making scheme(MBOMS)to minimize the long-term average weighted cost.The proposed MBOMS can learn the near-optimal offloading and migrating decision-making policy by centralized training and decentralized execution.Systematic and comprehensive simulation results reveal that our proposed MBOMS can converge well after training and outperforms the other five baseline algorithms.

Policy Network-Based Dual-Agent Deep Reinforcement Learning for Multi-Resource Task Offloading in Multi-Access Edge Cloud Networks

The Multi-access Edge Cloud(MEC) networks extend cloud computing services and capabilities to the edge of the networks. By bringing computation and storage capabilities closer to end-users and connected devices, MEC networks can support a wide range of applications. MEC networks can also leverage various types of resources, including computation resources, network resources, radio resources,and location-based resources, to provide multidimensional resources for intelligent applications in 5/6G.However, tasks generated by users often consist of multiple subtasks that require different types of resources. It is a challenging problem to offload multiresource task requests to the edge cloud aiming at maximizing benefits due to the heterogeneity of resources provided by devices. To address this issue,we mathematically model the task requests with multiple subtasks. Then, the problem of task offloading of multi-resource task requests is proved to be NP-hard. Furthermore, we propose a novel Dual-Agent Deep Reinforcement Learning algorithm with Node First and Link features(NF_L_DA_DRL) based on the policy network, to optimize the benefits generated by offloading multi-resource task requests in MEC networks. Finally, simulation results show that the proposed algorithm can effectively improve the benefit of task offloading with higher resource utilization compared with baseline algorithms.

Surveying on English Mobile Learning Among University Students: Current State and Influencing Factors

Mobile learning integrates mobile technology with digital learning,offering flexible,personalized content and portable equipment.It enables access to rich content and enhances learning efficiency.Therefore,it has become mainstream to utilize mobile devices for English learning among university students’English learning.The current study aims to examine the current situation and influencing factors of university students’English mobile learning.98 university students in one university of Shanghai participated the study and the questionnaire was used to collect the data.The results indicated that most university students already have electronic devices to support mobile learning.Personal factors,environmental factors,digital literacy,and technological capabilities are the main factors affecting university students’English mobile learning.The current study has implications for learners,teachers,and software developers.Learners should adjust their learning motivation,play an active role,and fully utilize the mobile platform to obtain resources and improve learning efficiency.Teachers should incorporate the advantages of mobile teaching and promote categorized and tiered teaching.Software developers should add new functions on the basis of meeting the basic needs of learners and continuously innovate the mobile learning platform.

Monitoring seismicity in the southern Sichuan Basin using a machine learning workflow

Monitoring seismicity in real time provides significant benefits for timely earthquake warning and analyses.In this study,we propose an automatic workflow based on machine learning(ML)to monitor seismicity in the southern Sichuan Basin of China.This workflow includes coherent event detection,phase picking,and earthquake location using three-component data from a seismic network.By combining Phase Net,we develop an ML-based earthquake location model called Phase Loc,to conduct real-time monitoring of the local seismicity.The approach allows us to use synthetic samples covering the entire study area to train Phase Loc,addressing the problems of insufficient data samples,imbalanced data distribution,and unreliable labels when training with observed data.We apply the trained model to observed data recorded in the southern Sichuan Basin,China,between September 2018 and March 2019.The results show that the average differences in latitude,longitude,and depth are 5.7 km,6.1 km,and 2 km,respectively,compared to the reference catalog.Phase Loc combines all available phase information to make fast and reliable predictions,even if only a few phases are detected and picked.The proposed workflow may help real-time seismic monitoring in other regions as well.

Energy-Efficient Traffic Offloading for RSMA-Based Hybrid Satellite Terrestrial Networks with Deep Reinforcement Learning

As the demands of massive connections and vast coverage rapidly grow in the next wireless communication networks, rate splitting multiple access(RSMA) is considered to be the new promising access scheme since it can provide higher efficiency with limited spectrum resources. In this paper, combining spectrum splitting with rate splitting, we propose to allocate resources with traffic offloading in hybrid satellite terrestrial networks. A novel deep reinforcement learning method is adopted to solve this challenging non-convex problem. However, the neverending learning process could prohibit its practical implementation. Therefore, we introduce the switch mechanism to avoid unnecessary learning. Additionally, the QoS constraint in the scheme can rule out unsuccessful transmission. The simulation results validates the energy efficiency performance and the convergence speed of the proposed algorithm.

Unified deep learning model for predicting fundus fluorescein angiography image from fundus structure image

The prediction of fundus fluorescein angiography(FFA)images from fundus structural images is a cutting-edge research topic in ophthalmological image processing.Prediction comprises estimating FFA from fundus camera imaging,single-phase FFA from scanning laser ophthalmoscopy(SLO),and three-phase FFA also from SLO.Although many deep learning models are available,a single model can only perform one or two of these prediction tasks.To accomplish three prediction tasks using a unified method,we propose a unified deep learning model for predicting FFA images from fundus structure images using a supervised generative adversarial network.The three prediction tasks are processed as follows:data preparation,network training under FFA supervision,and FFA image prediction from fundus structure images on a test set.By comparing the FFA images predicted by our model,pix2pix,and CycleGAN,we demonstrate the remarkable progress achieved by our proposal.The high performance of our model is validated in terms of the peak signal-to-noise ratio,structural similarity index,and mean squared error.

Efficacy of Online Learning and Influence on Skin and Eyes Among College Students

To evaluate the efficacy of online learning and explore the impact of long-term use of electronic products on facial skin as well as eyes.A cross-sectional survey was conducted to 180 sophomores in Xi′an Jiaotong University by cluster random sampling from September to October 2021.The questionnaire covering study condition,skin lesion and Ocular Surface Disease Index.χ_(2) test was used to compare the facial skin condition among different groups,and spearman correlation test was used to test the correlation of rank data.During online education,students′learning pressure is reduced,their autonomy is improved,and the learning efficiency is reduced.There were differences in the incidence of facial itching and papules among different groups.Duration of use of electronic products was positively correlated with the facial itching,with an r value of 0.231(P<0.05);the proportion of pigmentation in non-blue light protection groups(12.8%)was higher than that in blue light protection groups(1.7%),the difference was statistically significant(χ_(2)=8.384,P<0.05).The prevalence of dry eye among college students is 66.7%,and the proportion of moderate to severe dry eye is 34.5%.The study autonomy has been improved during online teaching.Long-term use of electronic products and no blue light protection have an impact on facial skin.Students should enhance the knowledge of skin-care and eye-care and develop better habits.

Influencing Factors of Chinese Proficiency Test Learning Application Under the Framework of Seamless Learning

With the development of modern information technology and network society,the era of digital transformation has arrived,many educational resources have been developed and utilized,and the problem of space-time separation of global Chinese learners is gradually being solved.The use of mobile application learning can achieve convenient and fast learning,reduce learning costs,and flexibly adjust learning progress.After experiencing COVID-19,the demand for human-computer interaction in mobile learning is even more necessary.Therefore,to make applications achieve seamless docking with the needs of learners in the design of learning,the research investigates and analyzes the five dimensions of time,space,mode,subject,and object.The research tests the reliability and validity of the questionnaire,removes the relevant items and conducts non-parametric tests,then puts forward corresponding strategies according to the results of the questionnaire.It aims to enhance the efficiency of Chinese learners in HSK(Hanyu Shuiping Kaoshi,Chinese Proficiency Test)learning and promote Chinese learning and the spread of the Chinese language and culture.

Research and Discussion on Flipped Classroom Combined with Case-Based Learning in Pharmacoeconomics Teaching

Objective:To explore the application effect of flipped classroom combined with case-based learning teaching methods in pharmacoeconomics teaching.Methods:The students majoring in clinical pharmacy in 2019 were selected as the study subjects,and the cost-effectiveness analysis of different dosage forms of Yinzhihuang in the treatment of neonatal jaundice was selected as the teaching case.The flipped classroom combined with case-based learning teaching method was used to carry out theoretical teaching to the students.After the course,questionnaires were distributed through the Sojump platform to evaluate the teaching effect.Results:The results of the questionnaire showed that 85.71%of the students believed that the flipped classroom combined with case-based learning teaching method was helpful in mobilizing the learning enthusiasm and initiative,and improving the comprehensive application ability of the knowledge of pharmacoeconomics.92.86%of the students think that it is conducive to the understanding and memorization of learning content,as well as the cultivation of teamwork,communication,etc.Conclusion:Flipped classroom combined with case-based learning teaching method can improve students’knowledge mastery,thinking skills,and practical application skills,as well as optimize and improve teachers’teaching levels.

Machine Learning‑Enhanced Flexible Mechanical Sensing

To realize a hyperconnected smart society with high productivity,advances in flexible sensing technology are highly needed.Nowadays,flexible sensing technology has witnessed improvements in both the hardware performances of sensor devices and the data processing capabilities of the device’s software.Significant research efforts have been devoted to improving materials,sensing mechanism,and configurations of flexible sensing systems in a quest to fulfill the requirements of future technology.Meanwhile,advanced data analysis methods are being developed to extract useful information from increasingly complicated data collected by a single sensor or network of sensors.Machine learning(ML)as an important branch of artificial intelligence can efficiently handle such complex data,which can be multi-dimensional and multi-faceted,thus providing a powerful tool for easy interpretation of sensing data.In this review,the fundamental working mechanisms and common types of flexible mechanical sensors are firstly presented.Then how ML-assisted data interpretation improves the applications of flexible mechanical sensors and other closely-related sensors in various areas is elaborated,which includes health monitoring,human-machine interfaces,object/surface recognition,pressure prediction,and human posture/motion identification.Finally,the advantages,challenges,and future perspectives associated with the fusion of flexible mechanical sensing technology and ML algorithms are discussed.These will give significant insights to enable the advancement of next-generation artificial flexible mechanical sensing.

Mechanism of Learning and Memory Impairment in Rats Exposed to Arsenic and/or Fluoride Based on Microbiome and Metabolome

Objective Arsenic(As) and fluoride(F) are two of the most common elements contaminating groundwater resources. A growing number of studies have found that As and F can cause neurotoxicity in infants and children, leading to cognitive, learning, and memory impairments. However, early biomarkers of learning and memory impairment induced by As and/or F remain unclear. In the present study, the mechanisms by which As and/or F cause learning memory impairment are explored at the multi-omics level(microbiome and metabolome).Methods We stablished an SD rats model exposed to arsenic and/or fluoride from intrauterine to adult period.Results Arsenic and/fluoride exposed groups showed reduced neurobehavioral performance and lesions in the hippocampal CA1 region. 16S rRNA gene sequencing revealed that As and/or F exposure significantly altered the composition and diversity of the gut microbiome, featuring the Lachnospiraceae_NK4A136_group, Ruminococcus_1, Prevotellaceae_NK3B31_group, [Eubacterium]_xylanophilum_group. Metabolome analysis showed that As and/or F-induced learning and memory impairment may be related to tryptophan, lipoic acid, glutamate, gamma-aminobutyric acidergic(GABAergic) synapse, and arachidonic acid(AA) metabolism. The gut microbiota, metabolites, and learning memory indicators were significantly correlated.Conclusion Learning memory impairment triggered by As and/or F exposure may be mediated by different gut microbes and their associated metabolites.

Task offloading mechanism based on federated reinforcement learning in mobile edge computing

With the arrival of 5G,latency-sensitive applications are becoming increasingly diverse.Mobile Edge Computing(MEC)technology has the characteristics of high bandwidth,low latency and low energy consumption,and has attracted much attention among researchers.To improve the Quality of Service(QoS),this study focuses on computation offloading in MEC.We consider the QoS from the perspective of computational cost,dimensional disaster,user privacy and catastrophic forgetting of new users.The QoS model is established based on the delay and energy consumption and is based on DDQN and a Federated Learning(FL)adaptive task offloading algorithm in MEC.The proposed algorithm combines the QoS model and deep reinforcement learning algorithm to obtain an optimal offloading policy according to the local link and node state information in the channel coherence time to address the problem of time-varying transmission channels and reduce the computing energy consumption and task processing delay.To solve the problems of privacy and catastrophic forgetting,we use FL to make distributed use of multiple users’data to obtain the decision model,protect data privacy and improve the model universality.In the process of FL iteration,the communication delay of individual devices is too large,which affects the overall delay cost.Therefore,we adopt a communication delay optimization algorithm based on the unary outlier detection mechanism to reduce the communication delay of FL.The simulation results indicate that compared with existing schemes,the proposed method significantly reduces the computation cost on a device and improves the QoS when handling complex tasks.

GrCol-PPFL:User-Based Group Collaborative Federated Learning Privacy Protection Framework

With the increasing number of smart devices and the development of machine learning technology,the value of users’personal data is becoming more and more important.Based on the premise of protecting users’personal privacy data,federated learning(FL)uses data stored on edge devices to realize training tasks by contributing training model parameters without revealing the original data.However,since FL can still leak the user’s original data by exchanging gradient information.The existing privacy protection strategy will increase the uplink time due to encryption measures.It is a huge challenge in terms of communication.When there are a large number of devices,the privacy protection cost of the system is higher.Based on these issues,we propose a privacy-preserving scheme of user-based group collaborative federated learning(GrCol-PPFL).Our scheme primarily divides participants into several groups and each group communicates in a chained transmission mechanism.All groups work in parallel at the same time.The server distributes a random parameter with the same dimension as the model parameter for each participant as a mask for the model parameter.We use the public datasets of modified national institute of standards and technology database(MNIST)to test the model accuracy.The experimental results show that GrCol-PPFL not only ensures the accuracy of themodel,but also ensures the security of the user’s original data when users collude with each other.Finally,through numerical experiments,we show that by changing the number of groups,we can find the optimal number of groups that reduces the uplink consumption time.

VPFL:A verifiable privacy-preserving federated learning scheme for edge computing systems

Federated learning for edge computing is a promising solution in the data booming era,which leverages the computation ability of each edge device to train local models and only shares the model gradients to the central server.However,the frequently transmitted local gradients could also leak the participants’private data.To protect the privacy of local training data,lots of cryptographic-based Privacy-Preserving Federated Learning(PPFL)schemes have been proposed.However,due to the constrained resource nature of mobile devices and complex cryptographic operations,traditional PPFL schemes fail to provide efficient data confidentiality and lightweight integrity verification simultaneously.To tackle this problem,we propose a Verifiable Privacypreserving Federated Learning scheme(VPFL)for edge computing systems to prevent local gradients from leaking over the transmission stage.Firstly,we combine the Distributed Selective Stochastic Gradient Descent(DSSGD)method with Paillier homomorphic cryptosystem to achieve the distributed encryption functionality,so as to reduce the computation cost of the complex cryptosystem.Secondly,we further present an online/offline signature method to realize the lightweight gradients integrity verification,where the offline part can be securely outsourced to the edge server.Comprehensive security analysis demonstrates the proposed VPFL can achieve data confidentiality,authentication,and integrity.At last,we evaluate both communication overhead and computation cost of the proposed VPFL scheme,the experimental results have shown VPFL has low computation costs and communication overheads while maintaining high training accuracy.

改进Q-Learning的路径规划算法研究

针对Q-Learning算法学习效率低、收敛速度慢且在动态障碍物的环境下路径规划效果不佳的问题,本文提出一种改进Q-Learning的移动机器人路径规划算法.针对该问题,算法根据概率的突变性引入探索因子来平衡探索和利用以加快学习效率;通过在更新函数中设计深度学习因子以保证算法探索概率;融合遗传算法,避免陷入局部路径最优同时按阶段探索最优迭代步长次数,以减少动态地图探索重复率;最后提取输出的最优路径关键节点采用贝塞尔曲线进行平滑处理,进一步保证路径平滑度和可行性.实验通过栅格法构建地图,对比实验结果表明,改进后的算法效率相较于传统算法在迭代次数和路径上均有较大优化,且能够较好的实现动态地图下的路径规划,进一步验证所提方法的有效性和实用性.

基于FL-MADQN算法的NR-V2X车载通信频谱资源分配

针对5G新空口-车联网(New Radio-Vehicle to Everything,NR-V2X)场景下车对基础设施(Vehicle to Infrastructure,V2I)和车对车(Vehicle to Vehicle,V2V)共享上行通信链路的频谱资源分配问题,提出了一种联邦-多智能体深度Q网络(Federated Learning-Multi-Agent Deep Q Network,FL-MADQN)算法.该分布式算法中,每个车辆用户作为一个智能体,根据获取的本地信道状态信息,以网络信道容量最佳为目标函数,采用DQN算法训练学习本地网络模型.采用联邦学习加快以及稳定各智能体网络模型训练的收敛速度,即将各智能体的本地模型上传至基站进行聚合形成全局模型,再将全局模型下发至各智能体更新本地模型.仿真结果表明:与传统分布式多智能体DQN算法相比,所提出的方案具有更快的模型收敛速度,并且当车辆用户数增大时仍然保证V2V链路的通信效率以及V2I链路的信道容量.