Federated Learning Model for Auto Insurance Rate Setting Based on Tweedie Distribution

In the assessment of car insurance claims,the claim rate for car insurance presents a highly skewed probability distribution,which is typically modeled using Tweedie distribution.The traditional approach to obtaining the Tweedie regression model involves training on a centralized dataset,when the data is provided by multiple parties,training a privacy-preserving Tweedie regression model without exchanging raw data becomes a challenge.To address this issue,this study introduces a novel vertical federated learning-based Tweedie regression algorithm for multi-party auto insurance rate setting in data silos.The algorithm can keep sensitive data locally and uses privacy-preserving techniques to achieve intersection operations between the two parties holding the data.After determining which entities are shared,the participants train the model locally using the shared entity data to obtain the local generalized linear model intermediate parameters.The homomorphic encryption algorithms are introduced to interact with and update the model intermediate parameters to collaboratively complete the joint training of the car insurance rate-setting model.Performance tests on two publicly available datasets show that the proposed federated Tweedie regression algorithm can effectively generate Tweedie regression models that leverage the value of data fromboth partieswithout exchanging data.The assessment results of the scheme approach those of the Tweedie regressionmodel learned fromcentralized data,and outperformthe Tweedie regressionmodel learned independently by a single party.

FedCLCC:A personalized federated learning algorithm for edge cloud collaboration based on contrastive learning and conditional computing

Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure challenges in edge environments.However,the diversity of clients in edge cloud computing presents significant challenges for FL.Personalized federated learning(pFL)received considerable attention in recent years.One example of pFL involves exploiting the global and local information in the local model.Current pFL algorithms experience limitations such as slow convergence speed,catastrophic forgetting,and poor performance in complex tasks,which still have significant shortcomings compared to the centralized learning.To achieve high pFL performance,we propose FedCLCC:Federated Contrastive Learning and Conditional Computing.The core of FedCLCC is the use of contrastive learning and conditional computing.Contrastive learning determines the feature representation similarity to adjust the local model.Conditional computing separates the global and local information and feeds it to their corresponding heads for global and local handling.Our comprehensive experiments demonstrate that FedCLCC outperforms other state-of-the-art FL algorithms.

Effects of mesenchymal stem cell on dopaminergic neurons,motor and memory functions in animal models of Parkinson's disease:a systematic review and meta-analysis

Parkinson’s disease is chara cterized by the loss of dopaminergic neurons in the substantia nigra pars com pacta,and although restoring striatal dopamine levels may improve symptoms,no treatment can cure or reve rse the disease itself.Stem cell therapy has a regenerative effect and is being actively studied as a candidate for the treatment of Parkinson’s disease.Mesenchymal stem cells are considered a promising option due to fewer ethical concerns,a lower risk of immune rejection,and a lower risk of teratogenicity.We performed a meta-analysis to evaluate the therapeutic effects of mesenchymal stem cells and their derivatives on motor function,memory,and preservation of dopamine rgic neurons in a Parkinson’s disease animal model.We searched bibliographic databases(PubMed/MEDLINE,Embase,CENTRAL,Scopus,and Web of Science)to identify articles and included only pee r-reviewed in vivo interve ntional animal studies published in any language through J une 28,2023.The study utilized the random-effect model to estimate the 95%confidence intervals(CI)of the standard mean differences(SMD)between the treatment and control groups.We use the systematic review center for laboratory animal expe rimentation’s risk of bias tool and the collaborative approach to meta-analysis and review of animal studies checklist for study quality assessment.A total of 33studies with data from 840 Parkinson’s disease model animals were included in the meta-analysis.Treatment with mesenchymal stem cells significantly improved motor function as assessed by the amphetamine-induced rotational test.Among the stem cell types,the bone marrow MSCs with neurotrophic factor group showed la rgest effect size(SMD[95%CI]=-6.21[-9.50 to-2.93],P=0.0001,I^(2)=0.0%).The stem cell treatment group had significantly more tyrosine hydroxylase positive dopamine rgic neurons in the striatum([95%CI]=1.04[0.59 to 1.49],P=0.0001,I^(2)=65.1%)and substantia nigra(SMD[95%CI]=1.38[0.89 to 1.87],P=0.0001,I^(2)=75.3%),indicating a protective effect on dopaminergic neurons.Subgroup analysis of the amphetamine-induced rotation test showed a significant reduction only in the intracranial-striatum route(SMD[95%CI]=-2.59[-3.25 to-1.94],P=0.0001,I^(2)=74.4%).The memory test showed significant improvement only in the intravenous route(SMD[95%CI]=4.80[1.84 to 7.76],P=0.027,I^(2)=79.6%).Mesenchymal stem cells have been shown to positively impact motor function and memory function and protect dopaminergic neurons in preclinical models of Parkinson’s disease.Further research is required to determine the optimal stem cell types,modifications,transplanted cell numbe rs,and delivery methods for these protocols.

大学生信息成瘾行为的触发路径与干预策略:基于S-O-R理论视角

[目的/意义]文章以大学生为研究对象,探究该群体信息成瘾行为的影响机制,为大学生信息成瘾行为的预防和管理提供价值参考。[方法/过程]引入刺激-机体-反应(S-O-R)理论,利用文献梳理总结大学生信息成瘾过程中的外部刺激,深度剖析机体在成瘾过程中产生的认知心理,构建大学生信息成瘾行为理论模型,运用结构方程模型实证分析大学生信息成瘾的触发路径。[结果/结论]研究结果表明,信息过载、间歇性奖励和行为管理会显著正向影响大学生的不确定性回避,进而导致大学生信息成瘾。同时,信息过载也会使大学生产生信息焦虑,信息焦虑显著正向影响大学生信息成瘾行为。面向高等院校、大学生群体和科技企业,提出针对性建议与对策,有助于完善高等院校管理体系,促进学生适应信息化环境,推动高等教育事业更好发展。

基于MCS-DCF模型的智慧交通企业价值评估

智慧交通行业具有综合性、系统性、数据驱动、服务智能化、安全性和环保性等特点,对相关企业进行价值评估可以为投资决策提供参考、为资产配置提供依据、为相关企业的并购重组提供指导、明确业务发展的方向等。在进行企业价值评估时,重要的是考虑当前实际情况,相应地调整使用的参数,选择恰当的评估方法,确保评估数据反映最新的市场动态和企业经营状况。文章以qf科技为例,使用灰色预测模型GM(1,1)对qf科技企业的未来自由现金流加以预测,通过DCF模型和经过蒙特卡洛模拟(Monte Carlo Simulation)改进后的DCF模型对qf科技企业价值进行评估,最终结果显示:通过蒙特卡洛模拟(Monte Carlo Simulation)改进后的模型评估企业价值更加合理。

Trusted Encrypted Traffic Intrusion Detection Method Based on Federated Learning and Autoencoder

With the rapid development of the Internet,network security and data privacy are increasingly valued.Although classical Network Intrusion Detection System(NIDS)based on Deep Learning(DL)models can provide good detection accuracy,but collecting samples for centralized training brings the huge risk of data privacy leakage.Furthermore,the training of supervised deep learning models requires a large number of labeled samples,which is usually cumbersome.The“black-box”problem also makes the DL models of NIDS untrustworthy.In this paper,we propose a trusted Federated Learning(FL)Traffic IDS method called FL-TIDS to address the above-mentioned problems.In FL-TIDS,we design an unsupervised intrusion detection model based on autoencoders that alleviates the reliance on marked samples.At the same time,we use FL for model training to protect data privacy.In addition,we design an improved SHAP interpretable method based on chi-square test to perform interpretable analysis of the trained model.We conducted several experiments to evaluate the proposed FL-TIDS.We first determine experimentally the structure and the number of neurons of the unsupervised AE model.Secondly,we evaluated the proposed method using the UNSW-NB15 and CICIDS2017 datasets.The exper-imental results show that the unsupervised AE model has better performance than the other 7 intrusion detection models in terms of precision,recall and f1-score.Then,federated learning is used to train the intrusion detection model.The experimental results indicate that the model is more accurate than the local learning model.Finally,we use an improved SHAP explainability method based on Chi-square test to analyze the explainability.The analysis results show that the identification characteristics of the model are consistent with the attack characteristics,and the model is reliable.

CCM-FL:Covert communication mechanisms for federated learning in crowd sensing IoT

The past decades have witnessed a wide application of federated learning in crowd sensing,to handle the numerous data collected by the sensors and provide the users with precise and customized services.Meanwhile,how to protect the private information of users in federated learning has become an important research topic.Compared with the differential privacy(DP)technique and secure multiparty computation(SMC)strategy,the covert communication mechanism in federated learning is more efficient and energy-saving in training the ma-chine learning models.In this paper,we study the covert communication problem for federated learning in crowd sensing Internet-of-Things networks.Different from the previous works about covert communication in federated learning,most of which are considered in a centralized framework and experimental-based,we firstly proposes a centralized covert communication mechanism for federated learning among n learning agents,the time complexity of which is O(log n),approximating to the optimal solution.Secondly,for the federated learning without parameter server,which is a harder case,we show that solving such a problem is NP-hard and prove the existence of a distributed covert communication mechanism with O(log logΔlog n)times,approximating to the optimal solution.Δis the maximum distance between any pair of learning agents.Theoretical analysis and nu-merical simulations are presented to show the performance of our covert communication mechanisms.We hope that our covert communication work can shed some light on how to protect the privacy of federated learning in crowd sensing from the view of communications.

Animal models of Alzheimer's disease:Current strategies and new directions

Animal models constructed using pathogenic factors have significantly advanced drug development for Alzheimer's disease(AD).These predominantly transgenic models,mainly in mice,replicate pathological phenotypes through gene mutations associated with familial AD cases,thus serving as vital tools for assessing drug efficacy and for performing mechanistic studies.However,the speciesspecific differences and complex,heterogeneous nature of AD etiology pose considerable challenges for the translatability of these animal models,limiting their utility in drug development.This review offers a comprehensive analysis of widely employed rodent(mice and rats)and non-rodent models(Danio rerio(zebrafish),Drosophila melanogaster,and Caenorhabditis elegans),detailing their phenotypic features and specific research applications.This review also examines the limitations inherent in these models and introduces various strategies for expanding AD modeling across diverse species,emphasizing recent advancement in non-human primates(NHPs)as valuable models.Furthermore,potential insights from the integration of innovative technologies in AD research are discussed,while providing valuable perspectives on the future development of AD animal models.

A Model of Grand Unified Theory: Suggested Solution for CP-Violation Using Ideas of Phase Paths

This study demonstrates that beyond standard model (BSM) cosmic fundamental interactions—weak, strong, and electromagnetic forces—can be unified through a common basis of representation. This unification allows for the derivation of the fine structure constant with running points of α(t) ≈ 1/(136.9038) at high energy scales, based on electroweak interactions. Through the application of the Ising model, the running point of the elementary charge e at high energy scales is determined, and Coulomb’s law is actually derived from the Yukawa potential. Theoretically, based on S. Weinberg’s electroweak interaction theory, this study unifies the strong and electromagnetic forces by representing them with rYuka, and further advances the reconstruction of the SU(3)C×SU(1)L×U(1)EMframework on the basis of electroweak interaction concepts. In fact, the cosmic fundamental forces can interchange at the mass gap, defined as the Yukawa turning phase at rYuka ≃1.9404 fm, with the SU(3)Diag structural constant fijk on glueballs calculated, estimating a spectrum mass gap of ∆0 > 0.

Evaluating the role of large language models in inflammatory bowel disease patient information

This letter evaluates the article by Gravina et al on ChatGPT’s potential in providing medical information for inflammatory bowel disease patients.While promising,it highlights the need for advanced techniques like reasoning+action and retrieval-augmented generation to improve accuracy and reliability.Emphasizing that simple question and answer testing is insufficient,it calls for more nuanced evaluation methods to truly gauge large language models’capabilities in clinical applications.

基于改进S-ResNet34模型的小麦条锈病等级识别研究

[目的]快速准确识别小麦条锈病病害等级,对其精准防控具有重要意义。[方法]利用数码相机获取小麦叶片条锈病RGB图像,构建小麦叶片条锈病不同病害等级数据集,通过对ResNet34模型添加通道注意力模块(SE)和Inception模块加以改进,增强模型对小麦条锈病特征的关注程度和提取能力,并采用精准率、召回率、平衡F分数和准确率等评价指标,对比分析S-ResNet34与VGG16、MobileNetV2、Swin-Transformer、ResNet34等多种主流模型的识别精度。[结果]S-ResNet34模型的训练准确率为93.85%,相比于VGG16(84.53%)、MobileNetV2(79.35%)、Swin-Transformer(85.67%)和ResNet34(87.50%)等深度网络模型,准确率分别提高了9.32%、14.50%、8.18%和6.35%。模型损失值更小,改进的ResNet34模型识别小麦条锈病特征能力更强,训练收敛更快。[结论]采用深度学习模型能够准确识别小麦条锈病发病程度,通过对ResNet34模型添加注意力模块能有效提高小麦条锈病病害识别精度。

Solving the Olbers’s Paradox, Explaining the “Red-Shift”, and Challenging the Relativities by “Sun Matters Theory” and “Sun Model of Universe”, an Evolution of the Einstein’s Static Universe Model

Olbers’s paradox, known as the dark night paradox, is an argument in astrophysics that the darkness of the night sky conflicts with the assumption of an infinite and eternal static universe. Big-Bang theory was used to partially explain this paradox, while introducing new problems. Hereby, we propose a better theory, named Sun Matters Theory, to explain this paradox. Moreover, this unique theory supports and extended the Einstein’s static universe model proposed by Albert Einstein in 1917. Further, we proposed our new universe model, “Sun Model of Universe”. Based on the new model and novel theory, we generated innovative field equation by upgrading Einstein’s Field Equation through adding back the cosmological constant, introducing a new variable and modifying the gravitationally-related concepts. According to the Sun Model of Universe, the dark matter and dark energy comprise the so-called “Sun Matters”. The observed phenomenon like the red shift is explained as due to the interaction of ordinary light with Sun Matters leading to its energy and frequency decrease. In Sun Model, our big universe consists of many universes with ordinary matter at the core mixed and surrounded with the Sun Matters. In those universes, the laws of physics may be completely or partially different from that of our ordinary universe with parallel civilizations. The darkness of night can be easily explained as resulting from the interaction of light with the Sun Matters leading to the sharp decrease in the light intensity. Sun Matters also scatter the light from a star, which makes it shining as observed by Hubble. Further, there is a kind of Sun Matters named “Sun Waters”, surrounding every starts. When lights pass by the sun, the Sun Waters deflect the lights to bend the light path. According to the Sun Model, it is the light bent not the space bent that was proposed in the theory of relativities.

Hubble Tension Explanation from This Cosmological Model AΛΩ (Slow Bang Model, SB)

In this article we present a model of Hubble-Lemaître law using the notions of a transmitter (galaxy) and a receiver (MW) coupled to a model of the universe (Slow Bang Model, SB), based on a quantum approach of the evolution of space-time as well as an equation of state that retains all the infinitesimal terms. We find an explanation of the Hubble tension H0. Indeed, we have seen that this constant depends on the transceiver pair which can vary from the lowest observable value, from photons of the CMB (theoretical [km/s/Mpc]) to increasingly higher values depending on the earlier origin of the formation of the observed galaxy or cluster (ETG ~0.3 [Gy], ~74 [km/s/Mpc]). We have produced a theoretical table of the values of the constant according to the possible pairs of transmitter/receiver in the case where these galaxies follow the Hubble flow without large disturbance. The calculated theoretical values of the constant are in the order of magnitude of all values mentioned in past studies. Subsequently, we applied the models to 9 galaxies and COMA cluster and found that the models predict acceptable values of their distances and Hubble constant since these galaxies mainly follow the Hubble flow rather than the effects of a galaxy cluster or a group of clusters. In conclusion, we affirm that this Hubble tension does not really exist and it is rather the understanding of the meaning of this constant that is questioned.

Analysis of Porter’s Five Forces Competitive Model in China’s Property Rights Trading Market

In recent years,China’s property rights trading market has been extremely competitive,but there are also new contradictions and challenges.This paper aims to analyze the external and internal competitive market of the property rights market through Porter’s five forces competition model,then find out the problems and defects in the development,thereby promoting the better development and progress of China’s property rights trading market.

Clustered Federated Learning with Weighted Model Aggregation for Imbalanced Data

As a promising edge learning framework in future 6G networks,federated learning(FL)faces a number of technical challenges due to the heterogeneous network environment and diversified user behaviors.Data imbalance is one of these challenges that can significantly degrade the learning efficiency.To deal with data imbalance issue,this work proposes a new learning framework,called clustered federated learning with weighted model aggregation(weighted CFL).Compared with traditional FL,our weighted CFL adaptively clusters the participating edge devices based on the cosine similarity of their local gradients at each training iteration,and then performs weighted per-cluster model aggregation.Therein,the similarity threshold for clustering is adaptive over iterations in response to the time-varying divergence of local gradients.Moreover,the weights for per-cluster model aggregation are adjusted according to the data balance feature so as to speed up the convergence rate.Experimental results show that the proposed weighted CFL achieves a faster model convergence rate and greater learning accuracy than benchmark methods under the imbalanced data scenario.

Exploring the role of N-acetyltransferases in diseases:a focus on N-acetyltransferase 9 in neurodegeneration

Acetyltransferases,required to transfer an acetyl group on protein are highly conserved proteins that play a crucial role in development and disease.Protein acetylation is a common post-translational modification pivotal to basic cellular processes.Close to 80%-90%of proteins are acetylated during translation,which is an irreversible process that affects protein structure,function,life,and localization.In this review,we have discussed the various N-acetyltransferases present in humans,their function,and how they might play a role in diseases.Furthermore,we have focused on N-acetyltransferase 9 and its role in microtubule stability.We have shed light on how N-acetyltransferase 9 and acetylation of proteins can potentially play a role in neurodegenerative diseases.We have specifically discussed the N-acetyltransferase 9-acetylation independent function and regulation of c-Jun N-terminal kinase signaling and microtubule stability during development and neurodegeneration.

Real-Time Prediction Algorithm for Intelligent Edge Networks with Federated Learning-Based Modeling

Intelligent healthcare networks represent a significant component in digital applications,where the requirements hold within quality-of-service(QoS)reliability and safeguarding privacy.This paper addresses these requirements through the integration of enabler paradigms,including federated learning(FL),cloud/edge computing,softwaredefined/virtualized networking infrastructure,and converged prediction algorithms.The study focuses on achieving reliability and efficiency in real-time prediction models,which depend on the interaction flows and network topology.In response to these challenges,we introduce a modified version of federated logistic regression(FLR)that takes into account convergence latencies and the accuracy of the final FL model within healthcare networks.To establish the FLR framework for mission-critical healthcare applications,we provide a comprehensive workflow in this paper,introducing framework setup,iterative round communications,and model evaluation/deployment.Our optimization process delves into the formulation of loss functions and gradients within the domain of federated optimization,which concludes with the generation of service experience batches for model deployment.To assess the practicality of our approach,we conducted experiments using a hypertension prediction model with data sourced from the 2019 annual dataset(Version 2.0.1)of the Korea Medical Panel Survey.Performance metrics,including end-to-end execution delays,model drop/delivery ratios,and final model accuracies,are captured and compared between the proposed FLR framework and other baseline schemes.Our study offers an FLR framework setup for the enhancement of real-time prediction modeling within intelligent healthcare networks,addressing the critical demands of QoS reliability and privacy preservation.

日本海沟S-net台网水平向海域浅地壳和上地幔地震动预测模型

地震动衰减关系是地震灾害研究和评估建筑结构可能遭受地震作用的一个重要工具.基于日本大型海底观测台网S-net的海域浅地壳和上地幔地震动记录,应用随机效应模型和回归分析构建了日本海沟地区的水平向海域浅地壳和上地幔地震动预测模型,对比了海域和陆地浅地壳和上地幔地震动预测模型在反应谱和衰减率上的差异.由于海底台站的埋设方式不同,在模型中将台站按埋设和未埋设进行区分,通过回归分析和参数平滑建立了考虑矩震级、断层距、震源深度、震源类型、水深、沉积层厚度和埋设方式的水平向海域浅地壳和上地幔地震动预测模型.通过与日本陆地预测模型的比较发现,海域地震动的长周期分量非常丰富,远大于陆域的预测值;在衰减率方面,短周期段低于陆地模型的衰减率.本文建立的海域浅地壳和上地幔地震动预测模型可为考虑震源分类的海域区划和地震危险性分析提供参考.

双酚S和双酚F诱导斑马鱼的神经发育毒性及联合作用分析

本研究选择亚致死剂量下等效应的BPS和BPF,从斑马鱼的发育表型、运动行为、血管发育和神经毒性等4个方面比较神经发育毒性的差异和联合毒性研究.结果表明,虽然LC5的BPS和BPF均会诱导斑马鱼表观畸形,如抑制了斑马鱼游泳囊、卵黄囊等发育;BPS和BPF的暴露会影响斑马鱼的早期运动行为,这与心血管系统损伤以及神经中枢系统损伤有着潜在的关联.BPS和BPF造成了斑马鱼心血管发育的异常,尾静脉(CV)宽度缩短;DA、AChE以及NO神经递质的异常传递导致了斑马鱼神经元发育受损,这对侧线神经系统的功能产生了直接的影响,进而导致斑马鱼早期运动行为的受损.BPF暴露更容易诱导斑马鱼的组织、器官毒性,而BPS直接毒性影响要小于BPF,更多的通过分子层次表现出来,这表明两者作用模式具有一定程度的差异性.此外,结合浓度加和模型的联合作用分析来分析20个检测指标,BPS和BPF产生加和作用有16个,达到了80%,因此混合物产生的毒性效应应当引起重视.研究结果为BPS、BPF及其混合物暴露后的健康风险评估与风险预警给予相关的参考依据.

Rock skeleton models and seismic porosity inversion

By substituting rock skeleton modulus expressions into Gassmann approximate fluid equation, we obtain a seismic porosity inversion equation. However, conventional rock skeleton models and their expressions are quite different from each other, resuling in different seismic porosity inversion equations, potentially leading to difficulties in correctly applying them and evaluating their results. In response to this, a uniform relation with two adjusting parameters suitable for all rock skeleton models is established from an analysis and comparison of various conventional rock skeleton models and their expressions including the Eshelby-Walsh, Pride, Geertsma, Nur, Keys-Xu, and Krief models. By giving the two adjusting parameters specific values, different rock skeleton models with specific physical characteristics can be generated. This allows us to select the most appropriate rock skeleton model based on geological and geophysical conditions, and to develop more wise seismic porosity inversion. As an example of using this method for hydrocarbon prediction and fluid identification, we apply this improved porosity inversion, associated with rock physical data and well log data, to the ZJ basin. Research shows that the existence of an abundant hydrocarbon reservoir is dependent on a moderate porosity range, which means we can use the results of seismic porosity inversion to identify oil reservoirs and dry or water-saturated reservoirs. The seismic inversion results are closely correspond to well log porosity curves in the ZJ area, indicating that the uniform relations and inversion methods proposed in this paper are reliable and effective.