FORECASTING NIKKEI 225 INDEX WITH SUPPORT VECTOR MACHINE

Support Vector Machine (SVM) is a very specific type of learning algorithms characterized by the capacity control of the decision function, the use of the kernel functions and the sparsity of the solution. In this paper, we investigate the predictability of financial movement direction with SVM by forecasting the weekly movement direction of NIKKEI 225 index. To evaluate the forecasting ability of SVM, we compare the perfor-mance with those of Linear Discriminant Analysis, Quadratic Discriminant Analysis and Elman Backpropagation Neural Networks. The experiment results show that SVM outperforms other classification methods. Furthermore, we propose a combining model by integrating SVM with other classification methods. The combining model performs the best among the forecasting methods.     

Inertial projection algorithms for convex feasibility problem

The purpose of this paper is to apply inertial technique to string averaging projection method and block-iterative projection method in order to get two accelerated projection algorithms for solving convex feasibility problem.Compared with the existing accelerated methods for solving the problem,the inertial technique employs a parameter sequence and two previous iterations to get the next iteration and hence improves the flexibility of the algorithm.Theoretical asymptotic convergence results are presented under some suitable conditions.Numerical simulations illustrate that the new methods have better convergence than the general projection methods.The presented algorithms are inspired by the inertial proximal point algorithm for finding zeros of a maximal monotone operator.     

A review and prospect of readable machine proofs for geometry theorems

After half a century research,the mechanical theorem proving in geometries has become an active research topic in the automated reasoning field.This review involves three approaches on automated generating readable machine proofs for geometry theorems which include search methods, coordinate-free methods,and formal logic methods.Some critical issues about these approaches are also discussed.Furthermore,the authors propose three further research directions for the readable machine proofs for geometry theorems,including geometry inequalities,intelligent geometry softwares and machine learning.     

Constrained voting extreme learning machine and its application

Extreme learning machine(ELM)has been proved to be an effective pattern classification and regression learning mechanism by researchers.However,its good performance is based on a large number of hidden layer nodes.With the increase of the nodes in the hidden layers,the computation cost is greatly increased.In this paper,we propose a novel algorithm,named constrained voting extreme learning machine(CV-ELM).Compared with the traditional ELM,the CV-ELM determines the input weight and bias based on the differences of between-class samples.At the same time,to improve the accuracy of the proposed method,the voting selection is introduced.The proposed method is evaluated on public benchmark datasets.The experimental results show that the proposed algorithm is superior to the original ELM algorithm.Further,we apply the CV-ELM to the classification of superheat degree(SD)state in the aluminum electrolysis industry,and the recognition accuracy rate reaches87.4%,and the experimental results demonstrate that the proposed method is more robust than the existing state-of-the-art identification methods.     

Self-adaptive learning based discrete differential evolution algorithm for solving CJWTA problem

Cooperative jamming weapon-target assignment （CJWTA） problem is a key issue in electronic countermeasures （ECM）. Some symbols which relevant to the CJWTA are defined firstly. Then, a formulation of jamming fitness is presented. Final y, a model of the CJWTA problem is constructed. In order to solve the CJWTA problem efficiently, a self-adaptive learning based discrete differential evolution （SLDDE） algorithm is proposed by introduc-ing a self-adaptive learning mechanism into the traditional discrete differential evolution algorithm. The SLDDE algorithm steers four candidate solution generation strategies simultaneously in the framework of the self-adaptive learning mechanism. Computa-tional simulations are conducted on ten test instances of CJWTA problem. The experimental results demonstrate that the proposed SLDDE algorithm not only can generate better results than only one strategy based discrete differential algorithms, but also outper-forms two algorithms which are proposed recently for the weapon-target assignment problems.     

Novel ensemble learning based on multiple section distribution in distributed environment

Because most ensemble learning algorithms use the centralized model, and the training instances must be centralized on a single station, it is difficult to centralize the training data on a station. A distributed ensemble learning algorithm is proposed which has two kinds of weight genes of instances that denote the global distribution and the local distribution. Instead of the repeated sampling method in the standard ensemble learning, non-balance sampling from each station is used to train the base classifier set of each station. The concept of the effective nearby region for local integration classifier is proposed, and is used for the dynamic integration method of multiple classifiers in distributed environment. The experiments show that the ensemble learning algorithm in distributed environment proposed could reduce the time of training the base classifiers effectively, and ensure the classify performance is as same as the centralized learning method.     

Improved artificial bee colony algorithm with mutual learning

The recently invented artificial bee colony (ABC) algorithm is an optimization algorithm based on swarm intelligence that has been used to solve many kinds of numerical function optimization problems.It performs well in most cases,however,there still exists an insufficiency in the ABC algorithm that ignores the fitness of related pairs of individuals in the mechanism of finding a neighboring food source.This paper presents an improved ABC algorithm with mutual learning (MutualABC) that adjusts the produced candidate food source with the higher fitness between two individuals selected by a mutual learning factor.The performance of the improved MutualABC algorithm is tested on a set of benchmark functions and compared with the basic ABC algorithm and some classical versions of improved ABC algorithms.The experimental results show that the MutualABC algorithm with appropriate parameters outperforms other ABC algorithms in most experiments.     

A review of addressing class noise problems of remote sensing classification

The development of image classification is one of the most important research topics in remote sensing. The prediction accuracy depends not only on the appropriate choice of the machine learning method but also on the quality of the training datasets. However, real-world data is not perfect and often suffers from noise. This paper gives an overview of noise filtering methods. Firstly, the types of noise and the consequences of class noise on machine learning are presented. Secondly, class noise ...     

Enhanced self-adaptive evolutionary algorithm for numerical optimization

There are many population-based stochastic search algorithms for solving optimization problems. However, the universality and robustness of these algorithms are still unsatisfactory. This paper proposes an enhanced self-adaptiveevolutionary algorithm (ESEA) to overcome the demerits above. In the ESEA, four evolutionary operators are designed to enhance the evolutionary structure. Besides, the ESEA employs four effective search strategies under the framework of the self-adaptive learning. Four groups of the experiments are done to find out the most suitable parameter values for the ESEA. In order to verify the performance of the proposed algorithm, 26 state-of-the-art test functions are solved by the ESEA and its competitors. The experimental results demonstrate that the universality and robustness of the ESEA out-perform its competitors.     

IDENTIFICATION OF TIME-VARIANT CONTINUOUS SYSTEM BY A QUASI-NEWTON METHOD

1 IntroductionThis paper presents a quasi-Newton method in infinite-dimensional spaces for identifyingparameters involved in a time-variant continuous, nonlinear system of differential equations.It is well known that quasi-Newton methods are considered to be those of the most success-ful algorithms used for numerically solving optimization problems in finite-dimensional spaces.But, few papers discuss these methods in system identification research. Because many iden-tification problems can be…     

New Hybrid Genetic Algorithm for Vertex Cover Problems

This paper presents a new hybrid genetic algorithm for the vertex cover problems in which scan-repair and local improvement techniques are used for local optimization. With the hybrid approach, genetic algorithms are used to perform global exploration in a population, while neighborhood search methods are used to perform local exploitation around the chromosomes. The experimental results indicate that hybrid genetic algorithms can obtain solutions of excellent quality to the problem instances with different sizes. The pure genetic algorithms are outperformed by the neighborhood search heuristics procedures combined with genetic algorithms.     

A rough set approach to feature selection based on scatter search metaheuristic

Rough set theory is an effective method to feature selection, which has recently fascinated many researchers. The essence of rough set approach to feature selection is to find a subset of the original features. It is, however, an NP-hard problem finding a minimal subset of the features, and it is necessary to investigate effective and efficient heuristic algorithms. This paper presents a novel rough set approach to feature selection based on scatter search metaheuristic. The proposed method, called scatter search rough set attribute reduction （SSAR）, is illustrated by 13 well known datasets from UCI machine learning repository. The proposed heuristic strategy is compared with typical attribute reduction methods including genetic algorithm, ant colony, simulated annealing, and Tabu search. Computational results demonstrate that our algorithm can provide efficient solution to find a minimal subset of the features and show promising and competitive performance on the considered datasets.     

Asian Option Pricing Based on Genetic Algorithms

The cross-fertilization between artificial intelligence and computational finance has resulted in some of the most active research areas in financial engineering. One direction is the application of machine learning techniques to pricing financial products, which is certainly one of the most complex issues in finance. In the literature, when the interest rate,the mean rate of return and the volatility of the underlying asset follow general stochastic processes, the exact solution is usually not available. In this paper, we shall illustrate how genetic algorithms (GAs), as a numerical approach, can be potentially helpful in dealing with pricing. In particular, we test the performance of basic genetic algorithms by using it to the determination of prices of Asian options, whose exact solutions is known from Black-Scholesoption pricing theory. The solutions found by basic genetic algorithms are compared with the exact solution, and the performance of GAs is ewluated accordingly. Based on these ewluations, some limitations of GAs in option pricing are examined and possible extensions to future works are also proposed.     

Class of modified parallel combined methods of real-time numerical simulation for a stiff system

A class of modified parallel combined methods of real-time numerical simulation are presented for a stiff dynamic system. By combining the parallelism across the system with the parallelism across the method, and relaxing the dependence of stage value computation on sampling time of input function, a class of modified real-time parallel combined methods are constructed. Stiff and nonstiff subsystems are solved in parallel on a parallel computer by a parallel Rosen-brock method and a parallel RK method, respectively. Their order conditions and convergences are discussed. The numerical simulation experiments show that this class of modified algorithms can get high speed and efficiency.     

Collaborative multi-agent reinforcement learning based on experience propagation

For multi-agent reinforcement learning in Markov games, knowledge extraction and sharing are key research problems. State list extracting means to calculate the optimal shared state path from state trajectories with cycles. A state list extracting algorithm checks cyclic state lists of a current state in the state trajectory, condensing the optimal action set of the current state. By reinforcing the optimal action selected, the action policy of cyclic states is optimized gradually. The state list extracting is repeatedly learned and used as the experience knowledge which is shared by teams. Agents speed up the rate of convergence by experience sharing. Competition games of preys and predators are used for the experiments. The results of experiments prove that the proposed algorithms overcome the lack of experience in the initial stage, speed up learning and improve the performance.     

PREDICTING CHAOTIC TIME SERIES WITH IMPROVED LOCAL APPROXIMATIONS

In this paper, new approaches for chaotic time series prediction are introduced. We first summarize and evaluate the existing local prediction models, then propose optimization models and new algorithms to modify procedures of local approximations. The modification to the choice of sample sets is given, and the zeroth-order approximation is improved by a linear programming method. Four procedures of first-order approximation are compared, and corresponding modified methods are given. Lastly, the idea of nonlinear feedback to raise predicting accuracy is put forward. In the end, we discuss two important examples, i.e. Lorenz system and Rossler system, and the simulation experiments indicate that the modified algorithms are effective.     

Novel algorithm for constructing support vector machine regression ensemble

1 .INTRODUCTIONRecently , support vector machine (SVM)[1]is anovel and promising technique in the fields of ma-chine learning and classification or regression pre-diction accompanying artificial neural network.InSVM,several learning algorithms can be obtainedgiven different inner-product functions named ker-nel functions ,such as polynomial approach,Bayes-ian classification、radial basic function method、multilayer perceptron network[2]. By now,it hasbeen successfully applied in many ar…     

Color-texture based unsupervised segmentation using JSEG with fuzzy connectedness

1.INTRODUCTIONJSEG[1]is a distinguished one in many existing seg-mentation methods based on color-texture[1,4,6~8].Itis computationally more feasible than many othermethods[1]because JSEGis to test the homogeneityof a given color-texture pattern,instead of doing pa-rameter esti mationfor a specific model like many oth-er texture segmentation algorithms.The segmentation results produced by JSEGaremainly based on a class-mapformed by the color classlabel of every pixel,which is the prod…     

Learning-based force servoing control of a robot with vision in an unknown environment

A learning-based control approach is presented for force servoing of a robot with vision in an unknown environment. Firstly, mapping relationships between image features of the servoing object and the joint angles of the robot are derived and learned by a neural network. Secondly, a learning controller based on the neural network is designed for the robot to trace the object. Thirdly, a discrete time impedance control law is obtained for the force servoing of the robot, the on-line learning algorithms for three neural networks are developed to adjust the impedance parameters of the robot in the unknown environment. Lastly, wiping experiments are carried out by using a 6 DOF industrial robot with a CCD camera and a force/torque sensor in its end effector, and the experimental results confirm the effecti veness of the approach.     

Radar emitter signal recognition method based on improved collaborative semi-supervised learning

Rare labeled data are difficult to recognize by using conventional methods in the process of radar emitter recognition. To solve this problem, an optimized cooperative semisupervised learning radar emitter recognition method based on a small amount of labeled data is developed. First, a small amount of labeled data are randomly sampled by using the bootstrap method, loss functions for three common deep learning networks are improved, the uniform distribution and cross-entropy function are combin...