Integration of large-scale cluster electric vehicles (EVs) and their spatial-temporal transfer randomness are likely to affect the safety and economic operation of the distribution network. This paper investigates the spatial-temporal distribution prediction of EVs’ charging load and then evaluates the reliability of the distribution network penetrated with large-scale cluster EVs. To effectively predict the charging load, trip chain technology, Monte Carlo method, and Markov decision process (MDP) theory are employed. Moreover, a spatial-temporal transfer model of EVs is established, and based on which, an EV energy consumption model and a charging load prediction model are constructed with consideration of temperature, traffic condition and EV owner’s subjective willingness in different scenarios. With the application of sequential Monte Carlo method, the paper further evaluates distribution network reliability in various charging scenarios. In the evaluation, indices including per unit value (PUV), fast voltage stability index (FVSI), loss of load probability (LOLP), system average interruption frequency index (SAIFI), system average interruption duration index (SAIDI), and expected energy not supplied (EENS) are incorporated. To validate the proposed prediction model and evaluation method, a series of numerical simulations are conducted on the basis of taking the traffic-distribution system of a typical city as an example. The result demonstrates that the proposed spatial-temporal transfer model is more practical in charging load prediction than the popularly used Dijkstra’s shortest path algorithm. Moreover, high temperature, congestion and the increment of EV penetration rate will further weaken distribution network reliability.
Federated learning (FL) is a promising technique for resolving the rising privacy and security concerns. Its main ingredient is to cooperatively learn the model among the distributed clients without uploading any sensitive data. In this article, we conducted a thorough review of the related works, following the development context and deeply mining the key technologies behind FL from the perspectives of theory and application. Specifically, we first classify the existing works in FL architecture based on the network topology of FL systems with detailed analysis and summarization. Next, we abstract the current application problems, summarize the general techniques, and frame the application problems into the general paradigm of FL base models. Moreover, we provide our proposed solutions for model training via FL. We have summarized and analyzed the existing FedOpt algorithms, and deeply revealed the algorithmic development principles of many first-order algorithms in depth, proposing a more generalized algorithm design framework. With the instantiation of these frameworks, FedOpt algorithms can be simply developed. As privacy and security are the fundamental requirements in FL, we provide the existing attack scenarios and the defense methods. To the best of our knowledge, we are among the first tier to review the theoretical methodology and propose our strategies since there are very few works surveying the theoretical approaches. Our survey targets motivating the development of high-performance, privacy-preserving, and secure methods to integrate FL into real-world applications.
31 December 2019 in IEEE Transactions on Neural Networks and Learning Systems
Ensuring the positive definiteness and avoiding ill conditioning of the Hessian update in the stochastic Broyden-Fletcher-Goldfarb-Shanno (BFGS) method are significant in solving nonconvex problems. This article proposes a novel stochastic version of a damped and regularized BFGS method for addressing the above problems. While the proposed regularized strategy helps to prevent the BFGS matrix from being close to singularity, the new damped parameter further ensures the positivity of the product of correction pairs. To alleviate the computational cost of the stochastic limited memory BFGS (LBFGS) updates and to improve its robustness, the curvature information is updated using the averaged iterate at spaced intervals. The effectiveness of the proposed method is evaluated through the logistic regression and Bayesian logistic regression problems in machine learning. Numerical experiments are conducted by using both synthetic data set and several real data sets. The results show that the proposed method generally outperforms the stochastic damped LBFGS (SdLBFGS) method. In particular, for problems with small sample sizes, our method has shown superior performance and is capable of mitigating ill-conditioned problems. Furthermore, our method is more robust to the variations of the batch size and memory size than the SdLBFGS method.
23 October 2017 in IEEE Transactions on Vehicular Technology
In this paper, we study the performance of the training based two-step channel estimation in two-way multipleinput multiple output (MIMO) relay systems over Rayleigh channels. We first estimate the forward channel during the first phase. During the second phase for the composite channel estimation, singular value decomposition (SVD) based maximum likelihood (ML) method is applied in two way relay systems, which is extended from one way relay systems. Hence, the backward channel is estimated with the given forward channel and composite channel. In general, the proposed estimation scheme can also be extended from the system with a single relay antenna to multiple relay antennas. Moreover, we derive the proposed training sequences and relay gain which aim to minimizes the trace of the approximate Bayesian Cramér-rao Lower Bound (ABCRLB) with respect to the individual channels. The numerical results are shown that although the training sequence is designed to improve individual channel MSE performance, it can also improve composite channel estimator performance.
6 April 2023 in ACM Transactions on Knowledge Discovery From Data
Federated optimization (FedOpt), which targets at collaboratively training a learning model across a large number of distributed clients, is vital for federated learning. The primary concerns in FedOpt can be attributed to the model divergence and communication efficiency, which significantly affect the performance. In this article, we propose a new method, i.e., LoSAC, to learn from heterogeneous distributed data more efficiently. Its key algorithmic insight is to locally update the estimate for the global full gradient after each regular local model update. Thus, LoSAC can keep clients’ information refreshed in a more compact way. In particular, we have studied the convergence result for LoSAC. Besides, the bonus of LoSAC is the ability to defend the information leakage from the recent technique Deep Leakage Gradients (DLG). Finally, experiments have verified the superiority of LoSAC comparing with state-of-the-art FedOpt algorithms. Specifically, LoSAC significantly improves communication efficiency by more than 100% on average, mitigates the model divergence problem, and equips with the defense ability against DLG.
This paper provides a detailed study on the convergence properties of the cubic regularized symmetric rank-1 (SR1) method (CuREG-SR1) proposed in [2] . The main advantage of incorporating cubic regularization technique to SR1 is to alleviate the problem of indefinite resulting matrix in SR1. However, its convergence under the line search framework is less studied. Here, we first show that CuREG-SR1 converges to a first-order critical point. Moreover, we give a novel result that provided the uniformly independent assumption, the difference between approximated Hessian matrix generated by CuREG-SR1 and the true Hessian is bounded. In addition, we show that for a problem with dimension $d$ , CuREG-SR1 generates $q-d$ superlinear steps every $q$ iterations. We also propose a novel incremental CuREG-SR1 (ICuREG-SR1) algorithm to adapt SR1 and CuREG-SR1 efficiently for solving large scale problems. The basic idea is to incorporate incremental optimization scheme, which updates progressively information for objective function involving a sum of individual functions, which are frequently encountered in large-scale machine learning. Numerical experiments on several machine learning problems show that the proposed algorithm offers superior performance in terms of the gradient magnitude than other conventional algorithms tested.