This paper proposes a multi-step prediction model for time series prediction, i.e. Meta-cognitive recurrent Kernel Online Sequential Extreme Learning Machine with Drift Detector Mechanism (Meta-RKOS-ELMALD). recurrent...
详细信息
This paper proposes a multi-step prediction model for time series prediction, i.e. Meta-cognitive recurrent Kernel Online Sequential Extreme Learning Machine with Drift Detector Mechanism (Meta-RKOS-ELMALD). recurrent multi-step algorithm is applied to release the limitation in the number of prediction steps, and Drift Detector Mechanism (DDM) is used to overcome the problem of concept drift in the prediction model. The new meta-cognitive strategy decides the way of the incoming data during training, which decreases the training computation of prediction model and solves the parameter dependency. In our evaluation, we use a total of six artificial data sets and three real-world data sets (Standard & Poor's 500 Index, Shanghai Stock Exchange Composite Index, and Ozone Concentration in Toronto) to prove the ability of kernel filters, the detecting ability of concept drift detector, and situation of applying meta-cognitive strategy in our proposed model. Experiments results indicate that the Meta-KOS-ELMALD with DDM has better forecasting ability in various predicting periods with the shortest learning time, as compared with other algorithms.
A method of real-time reconstruction of the useful signal and its lower derivatives on the basis of a recurrent smoothing spline is presented. A calculation technique for a spline with the number of measurements at ea...
详细信息
A method of real-time reconstruction of the useful signal and its lower derivatives on the basis of a recurrent smoothing spline is presented. A calculation technique for a spline with the number of measurements at each segment greater than the number of nodes is given, and the spline coefficients are found by the variational approach.
We propose recurrent algorithm for calculation of distribution of sum of exponentially distributed random values. algorithm can be used both for differently distributed random values and for coincidence of parameters ...
详细信息
We propose recurrent algorithm for calculation of distribution of sum of exponentially distributed random values. algorithm can be used both for differently distributed random values and for coincidence of parameters of certain random values.
New algorithms for computing power moments of hitting times and accumulated rewards of hitting type for semi-Markov processes are developed. The algorithms are based on special techniques of sequential phase space red...
详细信息
New algorithms for computing power moments of hitting times and accumulated rewards of hitting type for semi-Markov processes are developed. The algorithms are based on special techniques of sequential phase space reduction and recurrence relations connecting moments of rewards. Applications are discussed as well as possible generalizations of presented results and examples.
A problem of calculating the quality of the Triple Play service in an LTE mobile network under conditions of unicast and multicast transfer modes is considered. A mathematical model of the resource admission in the LT...
详细信息
A problem of calculating the quality of the Triple Play service in an LTE mobile network under conditions of unicast and multicast transfer modes is considered. A mathematical model of the resource admission in the LTE network is constructed in the form of a system with explicit losses and three disciplines of servicing the unicast, multicast, and elastic traffics. Approximate methods of calculating the stationary probability distribution of the states of the models with unicast and elastic traffics are discussed. An approximate method based on the calculation of the marginal distribution of the number of users with unicast and multicast traffics is proposed for calculating the average transfer time of the elastic traffic. An exact algorithm for determining the stationary probability distribution of the system's states that provides a considerable decrease in the dimension of the problem is presented.
A recurrent algorithm is devleloped for estimating hydroconductance coefficients taking into account errors of measurements of the input and output coordinates of a system. It is proposed to utilize algebraic polynomi...
详细信息
A recurrent algorithm is devleloped for estimating hydroconductance coefficients taking into account errors of measurements of the input and output coordinates of a system. It is proposed to utilize algebraic polynomials, splines, and crayging (a geostatistical method) in order to construct the hydroconductance field neglecting measurement noise, and to utilize a newly developed filter and smoothing splines taking measurement noise into account.
With the development of industry and technology, the development of the environment and cities has drawn lots of attention. Time series prediction plays a vital role in protecting the environment and improving the lev...
详细信息
With the development of industry and technology, the development of the environment and cities has drawn lots of attention. Time series prediction plays a vital role in protecting the environment and improving the level of intelligence and technology in cities, for example prediction of air pollution, water levels, palm oil prices, financial data and grid security. We describe a new algorithm, "Error-output recurrent Two-layer Extreme Learning Machine" or ERT-ELM: it applied a new recurrent technique, that not only removed the restriction of the prediction horizon problem, but it also used a mean squared error of the current step to update the output weights for the next step. This technique avoided error accumulation in the original recurrent algorithm for multi-step time series prediction. Moreover, the new two-layer structure network improved forecasting compared to conventional single-layer or two-layer ELM models. Quantum behaved Particle Swarm Optimization was used to find suitable ERT-ELM parameters. The ability of our model was assessed on ten data sets-two artificial and eight real-world data sets and performed significantly better than the baselines. Especially for the synthetic data sets, in 1-18 prediction periods, our model achieved mean square errors of 2.64 x 10(-3) on the Mackey-Glass data set and 1.49 x 10(-4) on the Lorenz data sets.
The purpose of the work is to design the algorithm that will allow to reduce the size of the collective dynamic routing optimal routes sets. The article contains the analysis of recurrent metaheuristicsbased algorithm...
详细信息
ISBN:
(纸本)9781538656235;9781538656228
The purpose of the work is to design the algorithm that will allow to reduce the size of the collective dynamic routing optimal routes sets. The article contains the analysis of recurrent metaheuristicsbased algorithm that was designed earlier. It is shown that it defines the routes sets that transmit more information than it is needed. That fact leads to the routes sets size and information delivery time increasing. The modified recurrent algorithm was designed. Its main feature is constraints of information to be transmitted. This feature allows to reduce the size of the optimal routes set and also the computational complexity of its definition procedure. In the MATLAB software environment, optimal routes sets size and computational complexity of the exact branch and bound method and recurrent metaheuristics-based algorithm are compared, as well as a modified recurrent algorithm. Obtained results showed that the introduction of the constraints in the modified algorithm makes it possible to approximate the size of the optimal routes sets to the solutions obtained by exact method and further reduce the computational complexity. Therefore designed algorithm is useful for decreasing the information delivery time and computational complexity.
The following paper discusses a conducted synthesis and assessment of the effectiveness of control algorithms for null formation in the radiation pattern of a flat phased antenna array according to two coordinates;the...
详细信息
ISBN:
(纸本)9781509010813
The following paper discusses a conducted synthesis and assessment of the effectiveness of control algorithms for null formation in the radiation pattern of a flat phased antenna array according to two coordinates;the procedure is performed by adjusting the amplitude distribution in rows and columns.
The minimum distance for linear codes is one of the important parameters. The shift bound is a good lower bound of the minimum distance for cyclic codes, Reed-Muller codes and geometric Goppa codes. It is necessary to...
详细信息
ISBN:
(纸本)9781467360500
The minimum distance for linear codes is one of the important parameters. The shift bound is a good lower bound of the minimum distance for cyclic codes, Reed-Muller codes and geometric Goppa codes. It is necessary to construct the maximum number of the independent set for the calculation of the shift bound. However, its computational complexity is very large, because the construction of the independent sets is not unique. The authors proposed an algorithm for calculation of the independent set and new lower bound using the discrete Fourier transform in 2010. In this paper we give simple modification and new recurrent algorithms to improve the original algorithm.
暂无评论