Objective Because it is impossible to know which statistical learning algorithm performs best on a prediction task, it is common to use stacking methods to ensemble individual learners into a more powerful single lear...
详细信息
Objective Because it is impossible to know which statistical learning algorithm performs best on a prediction task, it is common to use stacking methods to ensemble individual learners into a more powerful single learner. Stacking algorithms are usually based on linear models, which may run into problems, especially when predictions are highly correlated. In this study, we develop a greedy algorithm for model stacking that overcomes this issue while still being very fast and easy to interpret. We evaluate our greedy algorithm on 7 different data sets from various biomedical disciplines and compare it to linear stacking, genetic algorithm stacking and a brute force approach in different prediction settings. We further apply this algorithm on a task to optimize the weighting of the single domains (e.g., income, education) that build the German Index of Multiple Deprivation (GIMD) to be highly correlated with mortality. Results The greedy stacking algorithm provides good ensemble weights and outperforms the linear stacker in many tasks. Still, the brute force approach is slightly superior, but is computationally expensive. The greedy weighting algorithm has a variety of possible applications and is fast and efficient. A python implementation is provided.
The domain decomposition is a conventional approach to compute complex multidisciplinary simulations. The data transfer must be performed on their common interface due to the non-matching meshes for different domains....
详细信息
The domain decomposition is a conventional approach to compute complex multidisciplinary simulations. The data transfer must be performed on their common interface due to the non-matching meshes for different domains. The adaptive Kriging interpolation method is proposed based on the greedy algorithm. By introducing a learning function, the Kriging interpolation model is constructed by the source grid values. In this way, part of the total source points are selected and the Kriging model is of high precision for the data transfer between the interface. Four examples are investigated to demonstrate the efficiency and accuracy of the proposed method.
Phasor measurement units (PMUs) provide synchronized measurements of real-time phasors of voltages and currents. It is considered as an important element of the smart wide area measurement system used in advanced powe...
详细信息
Phasor measurement units (PMUs) provide synchronized measurements of real-time phasors of voltages and currents. It is considered as an important element of the smart wide area measurement system used in advanced power system monitoring, protection, and control applications. This paper proposes a new approach based on a greedy algorithm to solve the optimal phasor measurement unit placement (OPP) problem in the power network. The main purpose of proposed approach is to find out a high-quality solution in a reasonable time that ensures the practicability when applying for a real power network. The OPP problem is solved under both normal operating and contingency conditions. Moreover, some other realistic aspects that may affect the OPP problem, such as PMU channel limitation, zero injection bus, the presence of conventional measurements, are also considered to solve simultaneously. The simulations on IEEE 14-bus, 30-bus, 57-bus, 118-bus test systems, and especially on a large-scale network-the Polish 2383-bus system, are presented for evaluating the feasibility of proposed approach. The results of this study showed that the proposed method is effective and feasible to solve the OPP problem for a real power network.
We consider relationship between the rate of convergence to zero of the Faber-Schauder coefficients of continuous functions and the behavior of the greedy algorithm. We construct a continuous function f with Faber-Sch...
详细信息
We consider relationship between the rate of convergence to zero of the Faber-Schauder coefficients of continuous functions and the behavior of the greedy algorithm. We construct a continuous function f with Faber-Schauder coefficients |A(n)(f)| = O(log(-1)n) and divergent greedy algorithm.
Based on the Regularized Functional Matching Pursuit (RFMP) algorithm for linear inverse problems, we present an analogous iterative greedy algorithm for nonlinear inverse problems, called RFMP_NL. In comparison to es...
详细信息
Based on the Regularized Functional Matching Pursuit (RFMP) algorithm for linear inverse problems, we present an analogous iterative greedy algorithm for nonlinear inverse problems, called RFMP_NL. In comparison to established methods for nonlinear inverse problems, the algorithm is able to combine very diverse types of basis functions, for example, localized and global functions. This is important, in particular, in geoscientific applications, where global structures have to be distinguished from local anomalies. Furthermore, in contrast to other methods, the algorithm does not require the solution of large linear systems. We apply the RFMP_NL to the nonlinear inverse problem of gravimetry, where gravitational data are inverted for the shape of the surface or inner layer boundaries of planetary bodies. This inverse problem is described by a nonlinear integral operator, for which we additionally provide the Frechet derivative. Finally, we present two synthetic numerical examples to show that it is beneficial to apply the presented method to inverse gravimetric problems.
We present a systematic approach for salinity sensor placement in a polder network, where the objective is to estimate the unmeasured salinity levels in the main polder channels. We formulate this problem as optimizat...
详细信息
We present a systematic approach for salinity sensor placement in a polder network, where the objective is to estimate the unmeasured salinity levels in the main polder channels. We formulate this problem as optimization of the estimated salinity levels using root mean square error (RMSE) as the goodness of fit measure. Starting from a hydrodynamic and salt transport model of the Lissertocht catchment (a low-lying polder in the Netherlands), we use principal component analysis (PCA) to produce a low-order PCA model of the salinity distribution in the catchment. This model captures most of the relevant salinity dynamics and is capable of reconstructing the spatial and temporal salinity variation of the catchment. Just using three principal components (explaining 93% of the variance of the dataset) for the low-order PCA model, three optimally placed sensors with a greedy algorithm make the placement robust for modeling and measurement errors. The performance of the sensor placement for salinity reconstruction is evaluated against the detailed hydrodynamic and salt transport model and is shown to be close to the global optimum found by an exhaustive search with a RMSE of 82.2 mg/L.
In recent years, with the shortage of human resources and the increase in employment costs, the demand for intelligent processing system operations in various industries has increased. The smart car RGV can accept the...
详细信息
In recent years, with the shortage of human resources and the increase in employment costs, the demand for intelligent processing system operations in various industries has increased. The smart car RGV can accept the command to automatically control the moving direction, distance and realize the loading and unloading work, which greatly solves the problem of workshop work cost and efficiency. In this paper, the greedy algorithm is combined with MATLAB [1], JAVA [2] and other languages. According to the different RGV different workshop operations requirements, the corresponding RGV dynamic scheduling model and algorithm are established. The analytic hierarchy process is used to evaluate the model. Consistency test, finally by consulting the literature, the improved model is based on genetic algorithm and K-Means++ RGV dynamic scheduling model to reduce the system imbalance caused by CNC failure and meet greater production needs. For the task, we can divide the RGV operation into two specific situations: RGV processing of materials in one process: By analyzing the RGV straight-track round-trip loading and unloading operation, we made the total time of loading and unloading of 8 CNC machine tools (CNC) as the optimization goal, aiming to use the most time in the loading and unloading process within 8 hours of continuous operation shift. The figure describes the RGV straight-track operation. By comparing the start time and RGV arrival time of each CNC, a dynamic scheduling model based on TSP algorithm is established, and the iterative conditions are satisfied. The corresponding planning model is written and written in JAVA. Solve. The first group can process 146 materials. The occurrence of RGV failure and the elimination of the processing effects on the first and second processes: From the title, the probability of failure and the time of troubleshooting are known. The faulty machine number within the range and the exclusion time are generated by the random function.
We obtain Lebesgue-type inequalities for the greedy algorithm for arbitrary complete seminormalized biorthogonal systems in Banach spaces. The bounds are given only in terms of the upper democracy functions of the bas...
详细信息
We obtain Lebesgue-type inequalities for the greedy algorithm for arbitrary complete seminormalized biorthogonal systems in Banach spaces. The bounds are given only in terms of the upper democracy functions of the basis and its dual. We also show that these estimates are equivalent to embeddings between the given Banach space and certain discrete weighted Lorentz spaces. Finally, the asymptotic optimality of these inequalities is illustrated in various examples of not necessarily quasi-greedy bases.
In this paper, we prove that for any epsilon is an element of(0, 1) there exists ameasurable set E is an element of [0, 1) with measure vertical bar E vertical bar > 1 - epsilon such that for any function f is an e...
详细信息
In this paper, we prove that for any epsilon is an element of(0, 1) there exists ameasurable set E is an element of [0, 1) with measure vertical bar E vertical bar > 1 - epsilon such that for any function f is an element of L-1[0, 1), it is possible to construct a function (f) over tilde L1[0,1] coinciding with f on E and satisfying integral(1)(0)vertical bar(f) over tilde (x)-f(x)vertical bar dx greedy algorithm of <(f)over tilde> with respect to a bounded Vilenkin system are almost everywhere convergent on [0, 1).
Integrated radar and communication system can significantly enhance the work efficiency and considerably occupy less hardware resources. However, the integrated systems have limited supply resources, where a proper sc...
详细信息
Integrated radar and communication system can significantly enhance the work efficiency and considerably occupy less hardware resources. However, the integrated systems have limited supply resources, where a proper scheduling scheme is needed to meet the requirement of systems. Here, the authors propose a task-scheduling scheme based on the greedy algorithm, where each arrangement is the result of selecting the minimum shift time compared with the expected time. By guaranteeing the best result of each task, the minimum average time shift ratio (ATSR) of the whole tasks is obtained. Simulation results show that when the number of the tasks is quite large, the proposed scheme has more advantages than the traditional scheme based on time pointer algorithm. Authors' proposed scheme can significantly reduce the average offset time of tasks while maintaining a satisfactory successful scheduling rate (SSR), which shows more advantages in the integrated systems.
暂无评论