An Incentive Mechanism for Federated Learning:A Continuous Zero-Determinant Strategy Approach

As a representative emerging machine learning technique, federated learning(FL) has gained considerable popularity for its special feature of “making data available but not visible”. However, potential problems remain, including privacy breaches, imbalances in payment, and inequitable distribution.These shortcomings let devices reluctantly contribute relevant data to, or even refuse to participate in FL. Therefore, in the application of FL, an important but also challenging issue is to motivate as many participants as possible to provide high-quality data to FL. In this paper, we propose an incentive mechanism for FL based on the continuous zero-determinant(CZD) strategies from the perspective of game theory. We first model the interaction between the server and the devices during the FL process as a continuous iterative game. We then apply the CZD strategies for two players and then multiple players to optimize the social welfare of FL, for which we prove that the server can keep social welfare at a high and stable level. Subsequently, we design an incentive mechanism based on the CZD strategies to attract devices to contribute all of their high-accuracy data to FL.Finally, we perform simulations to demonstrate that our proposed CZD-based incentive mechanism can indeed generate high and stable social welfare in FL.

Improving Federated Learning through Abnormal Client Detection and Incentive

Data sharing and privacy protection are made possible by federated learning,which allows for continuous model parameter sharing between several clients and a central server.Multiple reliable and high-quality clients must participate in practical applications for the federated learning global model to be accurate,but because the clients are independent,the central server cannot fully control their behavior.The central server has no way of knowing the correctness of the model parameters provided by each client in this round,so clients may purposefully or unwittingly submit anomalous data,leading to abnormal behavior,such as becoming malicious attackers or defective clients.To reduce their negative consequences,it is crucial to quickly detect these abnormalities and incentivize them.In this paper,we propose a Federated Learning framework for Detecting and Incentivizing Abnormal Clients(FL-DIAC)to accomplish efficient and security federated learning.We build a detector that introduces an auto-encoder for anomaly detection and use it to perform anomaly identification and prevent the involvement of abnormal clients,in particular for the anomaly client detection problem.Among them,before the model parameters are input to the detector,we propose a Fourier transform-based anomaly data detectionmethod for dimensionality reduction in order to reduce the computational complexity.Additionally,we create a credit scorebased incentive structure to encourage clients to participate in training in order tomake clients actively participate.Three training models(CNN,MLP,and ResNet-18)and three datasets(MNIST,Fashion MNIST,and CIFAR-10)have been used in experiments.According to theoretical analysis and experimental findings,the FL-DIAC is superior to other federated learning schemes of the same type in terms of effectiveness.

A multi-scale second-order autoregressive recursive filter approach for the sea ice concentration analysis

To effectively extract multi-scale information from observation data and improve computational efficiency,a multi-scale second-order autoregressive recursive filter(MSRF)method is designed.The second-order autoregressive filter used in this study has been attempted to replace the traditional first-order recursive filter used in spatial multi-scale recursive filter(SMRF)method.The experimental results indicate that the MSRF scheme successfully extracts various scale information resolved by observations.Moreover,compared with the SMRF scheme,the MSRF scheme improves computational accuracy and efficiency to some extent.The MSRF scheme can not only propagate to a longer distance without the attenuation of innovation,but also reduce the mean absolute deviation between the reconstructed sea ice concentration results and observations reduced by about 3.2%compared to the SMRF scheme.On the other hand,compared with traditional first-order recursive filters using in the SMRF scheme that multiple filters are executed,the MSRF scheme only needs to perform two filter processes in one iteration,greatly improving filtering efficiency.In the two-dimensional experiment of sea ice concentration,the calculation time of the MSRF scheme is only 1/7 of that of SMRF scheme.This means that the MSRF scheme can achieve better performance with less computational cost,which is of great significance for further application in real-time ocean or sea ice data assimilation systems in the future.

Coordinated Control of Second-order Harmonic Voltage and Current Fluctuations in Cascaded H-bridge Inverter with Supercapacitor and DC-DC Stage at Variable Output Frequencies

In the cascaded H-bridge inverter(CHBI)with supercapacitor and dc-dc stage,inherent second-order harmonic power flows through each submodule(SM),causing fluctuations in both the dc-link voltage and the dc-dc current.There exist limitations in handling these fluctuations at variable output frequencies when employing proportional-integral(PI)control to the dc-dc stage.This paper aims to coordinately control these second-order harmonic voltage and current fluctuations in the CHBI.The presented method configures a specific second-order harmonic voltage reference,equipped with a maximum voltage fluctuation constraint and a suitable phase,for the dc-dc stage.A PI-resonant controller is used to track the configured reference.This allows for regulating the second-order harmonic fluctuation in the average dc-link voltage among the SMs within a certain value.Importantly,the second-order harmonic fluctuation in the dc-dc current can also be reduced.Simulation and experimental results demonstrate the effectiveness of the presented method.

Carbon Abatement Cost-Sharing Strategy for Electric Power Sector Based on Incentive and Subsidy Mechanisms

The green and low carbon transition and development of the electricity industry is the most crucial task in realizing the“dual-carbon target”,and it is urgent to explore the incentive and subsidy mechanism to promote green electricity consumption and the cost-sharing strategy of carbon reduction,to alleviate the pressure of carbon abatement cost of each subject of the electricity supply chain.Against this background,this paper takes into account the low-carbon subsidies provided by the government and the incentive subsidies for users,and studies the optimal decision-making of each subject in the electricity supply chain,so that each of them can obtain the optimal profit and achieve carbon emission reduction at the same time.Firstly,taking into account the direct power purchase mode of large users and the electricity-selling companies emerging after the reform of the power sales side,we have established a cooperative mechanism for sharing the cost of carbon emission reduction in the electricity supply chain and clarified the relationship between the supply and demand of electricity among the main parties.Subsequently,considering government low-carbon subsidies and user incentive subsidies,the optimal decisionmaking model is established under two scenarios of decentralized and centralized cooperative games in the supply chain,respectively,with the objective of maximizing profits and carbon reduction rates.Solving for the optimal proportion of carbon abatement costs shared by each participant in the electricity supply chain in achieving game equilibrium.Finally,we analyze the role of the government’s low-carbon subsidies,users’incentive subsidies,and other factors on the profit and carbon reduction effect of the electricity industry through the example analysis and further analyze the impact of carbon abatement cost-sharing measures to provide recommendations for the electricity industry to realize low-carbon abatement and make decisions.

Stackelberg Game-Based Optimal Dispatch for PEDF Park and Power Grid Interaction under Multiple Incentive Mechanisms

The integration of photovoltaic,energy storage,direct current,and flexible load(PEDF)technologies in building power systems is an importantmeans to address the energy crisis and promote the development of green buildings.The friendly interaction between the PEDF systems and the power grid can promote the utilization of renewable energy and enhance the stability of the power grid.For this purpose,this work introduces a framework of multiple incentive mechanisms for a PEDF park,a building energy system that implements PEDF technologies.The incentive mechanisms proposed in this paper include both economic and noneconomic aspects,which is the most significant innovation of this paper.By modeling the relationship between a PEDF park and the power grid into a Stackelberg game,we demonstrate the effectiveness of these incentive measures in promoting the friendly interaction between the two entities.In this game model,the power grid determines on the prices of electricity trading and incentive subsidy,aiming to maximize its revenue while reducing the peak load of the PEDF park.On the other hand,the PEDF park make its dispatch plan according to the prices established by the grid,in order to reduce electricity consumption expense,improve electricity utility,and enhance the penetration rate of renewable energy.The results show that the proposed incentive mechanisms for the PEDF park can help to optimize energy consumption and promote sustainable energy practices.

Flexible Load Participation in Peaking Shaving and Valley Filling Based on Dynamic Price Incentives

Considering the widening of the peak-valley difference in the power grid and the difficulty of the existing fixed time-of-use electricity price mechanism in meeting the energy demand of heterogeneous users at various moments or motivating users,the design of a reasonable dynamic pricing mechanism to actively engage users in demand response becomes imperative for power grid companies.For this purpose,a power grid-flexible load bilevel model is constructed based on dynamic pricing,where the leader is the dispatching center and the lower-level flexible load acts as the follower.Initially,an upper-level day-ahead dispatching model for the power grid is established,considering the lowest power grid dispatching cost as the objective function and incorporating the power grid-side constraints.Then,the lower level comprehensively considers the load characteristics of industrial load,energy storage,and data centers,and then establishes a lower-level flexible load operation model with the lowest user power-consuming cost as the objective function.Finally,the proposed method is validated using the IEEE-118 system,and the findings indicate that the dynamic pricing mechanism for peaking shaving and valley filling can effectively guide users to respond actively,thereby reducing the peak-valley difference and decreasing users’purchasing costs.

Research on the Improvement of Incentive Mechanism for Pharmaceutical Sales Personnel - Taking Company A as an Example

Objective To analyze the improvement of the incentive mechanism of sales personnel in pharmaceutical company A,and to promote the smooth operation and further development of the company in a long term.Methods Compensation incentive,performance appraisal,welfare benefit,training incentive,promotion motivation and enterprise cultural inspiration were explored through questionnaires,telephone interviews and in-person interviews.Results and Conclusion This company’s incentive mechanism has problems in two aspects:Material incentives and spiritual incentives.As to the company’s characteristics and strategic development,the optimization countermeasures of incentive mechanism are proposed from the following three aspects:constructing a reasonable incentive system,establishing an efficient spiritual incentive mechanism,and implementing the dynamic incentive and differentiated incentive simultaneously.

Will the perceptions to incentive policies have any effect on users' willingness to participate in Mobility-as-a-Service?

With the increasing severity of urban traffic congestion and environmental pollution issues,Mobility-as-a-Service(MaaS)has garnered increasing attention as an emerging mode of transportation.Thus,how to motivate users to participate in MaaS has become an important research issue.This study first classified the incentive policies into four aspects:financial incentive policy,non-financial incentive policy,information policy,and convenience policy.Then,through online questionnaires and field interviews,456 sets of data were collected in Beijing,and the data were analyzed by the structural equation model and latent class model.The results show that the four incentive policies are positively correlated with users'participation in MaaS,among which financial incentive policy and information policy have the greatest impact,that is,they can better encourage users by increasing direct financial subsidies and broadening the information about MaaS.In addition,Latent Class Analysis was performed to class different users and it was found that the personal characteristics of users had some influence on willingness to participate in MaaS.Therefore,incentive policies should be designed to consider the needs and characteristics of different user groups to improve their willingness to participate in MaaS.The results can provide theoretical suggestions for the government to promote the widespread application of MaaS in urban transportation.

Research on Career Development Paths and Incentive Model Design for Middle Managers in Banks

With the rapid development of financial technology,middle managers in banks face both new challenges and opportunities.This paper conducts an in-depth analysis of the current state and issues surrounding the career development of middle managers in banks and explores effective incentive model designs and implementation strategies.Employing both quantitative and qualitative evaluation methods,the study assesses the practical impact of various incentive models.Through case analysis,targeted improvement suggestions are proposed.The findings reveal that a well-designed incentive mechanism significantly enhances middle managers’job satisfaction and loyalty,which is essential for the sustained growth of banks.

The Convergence Rate of Fréchet Distribution under the Second-Order Regular Variation Condition

In this article we consider the asymptotic behavior of extreme distribution with the extreme value index γ>0 . The rates of uniform convergence for Fréchet distribution are constructed under the second-order regular variation condition.

Incentive mechanism analysis of information security outsourcing based on principal-agent model

In order to solve principal-agent problems caused by interest inconformity and information asymmetry during information security outsourcing, it is necessary to design a reasonable incentive mechanism to promote client enterprises to complete outsourcing service actively. The incentive mechanism model of information security outsourcing is designed based on the principal-agent theory. Through analyzing the factors such as enterprise information assets value, invasion probability, information security environment, the agent cost coefficient and agency risk preference degree how to impact on the incentive mechanism, conclusions show that an enterprise information assets value and invasion probability have a positive influence on the fixed fee and the compensation coefficient; while information security environment, the agent cost coefficient and agency risk preference degree have a negative influence on the compensation coefficient. Therefore, the principal enterprises should reasonably design the fixed fee and the compensation coefficient to encourage information security outsourcing agency enterprises to the full extent.

Second-Order Raman Scattering from n- and p-Type 4H-SiC

The results of second-order Raman-scattering experiments on n- and p-type 4H-SiC are presented,covering the acoustic and the optical overtone spectral regions.Some of the observed structures in the spectra are assigned to particular phonon branches and the points in the Brillouin zone from which the scattering originates.There exists a doublet at 626/636cm -1 with energy difference about 10cm -1 in both n- and p-type 4H-SiC,which is similar to the doublet structure with the same energy difference founded in hexagonal GaN,ZnO, and AlN.The cutoff frequency at 1926cm -1 of the second-order Raman is not the overtone of the A 1(LO) peak of the n-type doping 4H-SiC,but that of the undoping one.The second-order Raman spectrum of 4H-SiC can hardly be affected by doping species or doping density.

Anti-periodic solutions to a class of second-order evolution equations

In this paper we discuss the anti-periodic problem for a class of abstractnonlinear second-order evolution equations associated with maximal monotone operators in Hilbertspaces and give some new assumptions on operators. We establish the existence and uniqueness ofanti-periodic solutions, which improve andgeneralize the results that have been obtained. Finally weillustrate the abstract theory by discussing a simple example of an anti-periodic problem fornonlinear partial differential equations.

Anti-interference beam pattern design based on second-order cone programming optimization

When signal-to-interference ratio is low, the energy of strong interference leaked from the side lobe of beam pattern will infect the detection of weak target. Therefore, the beam pattern needs to be optimized. The existing Dolph-Chebyshev weighting method can get the lowest side lobe level under given main lobe width, but for the other non-uniform circular array and nonlinear array, the low side lobe pattern needs to be designed specially. The second order cone programming optimization （SOCP） algorithm proposed in the paper transforms the optimization of the beam pattern into a standard convex optimization problem. Thus there is a paradigm to follow for any array formation, which not only achieves the purpose of Dolph-Chebyshev weighting, but also solves the problem of the increased side lobe when the signal is at end fire direction The simulation proves that the SOCP algorithm can detect the weak target better than the conventional beam forming.

A Study on the Multi-Task Incentive Mechanism for the Service Provider

Both conflict and asymmetric information exist betweenthe telecom operators and the service provider,and result in illegal behaviors of the service provider.The relationship between the telecom operators andthe service provider is classical multi-task principalagentrelationship. The multi-task incentive for theservice provider is considered in the design of theprincipal-agent incentive contract, and it is necessaryto add the multi-task incentive to the serviceproviders through the analysis of the risk costs andthe agency costs of this problem.

RTRC:A Reputation-Based Incentive Game Model for Trustworthy Crowdsourcing Service

The ubiquity of mobile devices have promoted the prosperity of mobile crowd systems, which recruit crowds to contribute their resources for performing tasks. Yet, due to the various resource consumption, the crowds may be reluctant to join and contribute information. Thus, the low participation level of crowds will be a hurdle that prevents the adoption of crowdsourcing. A critical challenge for these systems is how to design a proper mechanism such that the crowds spontaneously act as suppliers to contribute accurate information. Most of existing mechanisms ignore either the honesty of crowds or requesters respectively. In this paper, considering the honesty of both, we propose a game-based incentive mechanism, namely RTRC, to stimulate the crowds to contribute accurate information and to motivate the requesters to return accurate feedbacks. In addition, an evolutionary game is designed to model the dynamic of user-strategy selection. Specially, the replicator dynamic is applied to model the adaptation of strategy interactions taking into account the dynamic nature in time dependence and we also derive the evolutionarily stable strategies(ESSs) for users. Finally, empirical results over the simulations show that all the requesters and suppliers will select honest strategy to maximize their profit.

Stability analysis for a second-order continuous finite-time control system subject to a disturbance

In this paper, using finite-time control method, we consider the disturbance analysis of a second-order system with unknown but bounded disturbance. We show that the states of the second-order system will be stabilized to a region containing the origin. The radius of this region is determined by the control parameters and can be rendered as small as desired. The rigorous stability analysis is also given. Compared with the conventional PD control law, the finite-time control law yields a better disturbance rejection performance. Numerical simulation results show the effectiveness of the method.

Incentive Scheme for Slice Cooperation Based on D2D Communication in 5G Networks

In the 5th generation(5G)wireless communication networks,network slicing emerges where network operators(NPs)form isolated logical slices by the same cellular network infrastructure and spectrum resource.In coverage regions of access points(APs)shared by slices,device to device(D2D)communication can occur among different slices,i.e.,one device acts as D2D relay for another device serving by a different slice,which is defined as slice cooperation in this paper.Since selfish slices will not help other slices by cooperation voluntarily and unconditionally,this paper designs a novel resource allocation scheme to stimulate slice cooperation.The main idea is to encourage slice to perform cooperation for other slices by rewarding it with higher throughput.The proposed incentive scheme for slice cooperation is formulated by an optimal problem,where cooperative activities are introduced to the objective function.Since optimal solutions of the formulated problem are long term statistics,though can be obtained,a practical online slice scheduling algorithm is designed,which can obtain optimal solutions of the formulated maximal problem.Lastly,the throughput isolation indexes are defined to evaluate isolation performance of slice.According to simulation results,the proposed incentive scheme for slice cooperation can stimulate slice cooperation effectively,and the isolation of slice is also simulated and discussed.