The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Here I apply three reinforcement learning methods to the full, continuous action, swing-up acrobot control benchmark problem. These include two approaches from the literature: CACLA and NM-SARSA and a novel approach which I refer to as Nelder Mead-SARSA. Nelder Mead-SARSA, like NMSARSA, directly optimises the state-action value function for action selection, in order to allow continuous action reinforcement...
This paper presents a novel algorithm for multiobjective training of Radial Basis Function (RBF) networks based on least-squares and Particle Swarm Optimization methods. The formulation is based on the fundamental concept that supervised learning is a bi-objective optimization problem, in which two conflicting objectives should be minimized. The objectives are related to the empirical training error...
Since the excellent performances of treating nonlinear data with self-learning capability, the neural networks (NNs) are wildly use in financial prediction problem. But the NNs more or less suffer from the slow convergence, “black-box” i.e., it is almost impossible to analysis them for how they work. The Fuzzy Neural Networks(FNN) allow to add rules to neural networks. This avoids the black-box but...
Extension neural network is a new method based on Extenics and neural networks, it is full use of the Extension of qualitative and quantitative description of the advantages, but also consider the parallel structure characteristics, of neural network. This article describes the extension theory and neural network fusion extension neural network structure and introduce ENN algorithm based on genetic...
Ensemble learning is a method to improve the performance of classification and prediction algorithms. It has received considerable attention because of its prominent generalization and performance improvement. However, its performance can be degraded due to multicollinearity problem where multiple classifiers of an ensemble are highly correlated with. This paper proposes genetic algorithm-based coverage...
In this paper, we propose a cooperative learning algorithm for Multi-category classification which is decomposed into two sub-optimization problems by using the support vector machine technique. The proposed cooperative learning algorithm consists of two single learning algorithms and each sub-optimization problem is solved by one of them. Unlike the cooperative neural network, the proposed cooperative...
A key feature in population based optimization algorithms is the ability to explore a search space and make a decision based on multiple solutions. In this paper, an incremental learning strategy based on a dynamic particle swarm optimization (DPSO) algorithm allows to produce heterogeneous ensembles of classifiers for video-based face recognition. This strategy is applied to an adaptive classification...
In order to improve the generalization ability of feed-forward neural networks, a new objective function of learning procedure for training single hidden layer network is proposed. This objective function is composed of two information entropy, one is the cross entropy as the main optimization term and the other is the fuzzy entropy as the regularization term. In this paper, we are fused the concept...
In this paper, a supervised learning strategy based on a Multi-Objective Particle Swarm Optimization (MOPSO) is introduced for ARTMAP neural networks. It is based on the concept of neural network evolution in that particles of a MOPSO swarm (i.e., network solutions) seek to determine user-defined parameters and network (weights and architecture) such that generalisation error and network resources...
A evolutionary programming is proposed in this paper to automatically design neural networks(NNS) ensembles. Based on negative correlation learning, different individual NNs in the ensemble can learn to subdivide the task and thereby solve it more efficiently and elegantly. At the same time, different individual NNs are always to find the best collaboration connection during the evolutionary process...
This paper considers on-line training of feadforward neural networks. Training examples are only available sampled randomly from a given generator. What emerges in this setting is the problem of step-sizes, or learning rates, adaptation. A scheme of determining step-sizes is introduced here that satisfies the following requirements: (i) it does not need any auxiliary problem-dependent parameters,...
The paper discusses the quadratic neural unit (QNU) and highlights its attractiveness for industrial applications such as for plant modeling, control, and time series prediction. Linear systems are still often preferred in industrial control applications for their solvable and single solution nature and for the clarity to the most application engineers. Artificial neural networks are powerful cognitive...
The Support Vector Machines (SVM) become popular E-Business data mining tools recently, and the datasets of E-Business are usually large-scale. If Support Vector Machines are trained on large-scale datasets, the training time will be very long and the classifier's accuracy will become lower too. As training a large-scale SVM is equated to solve a large-scale quadratic programming (QP) problem, so...
Imperialist Competitive Algorithm (ICA) is a novel optimization algorithm that inspired by socio-political process of imperialistic competition. ICA shown its excellent capability in diverse optimization tasks. In this paper, a new method for training an Artificial Neural Network using Chaotic Imperialist Competitive Algorithm is proposed. In Chaotic Imperialist Competitive Algorithm (CICA) the chaos...
To address training of process neural networks based on the orthogonal basis expansion, a double chains quantum genetic algorithm based on the probability amplitudes of quantum bits is proposed. In this method, the probability amplitudes of each qubit are regarded as two paratactic genes, each chromosome contains two gene chains, and each of gene chains represents an optimization solution. The number...
Propose a wavelet neural network (WNN) sound source model based on Shuffled Frog Leaping Algorithm (SFLA). Utilize frog leaping algorithm to optimize weights and thresholds of WNN, obtain initial weights and thresholds possessing certain ergodicity and then train WNN. It overcomes disadvantages of neural network that has slower searching speed and easily falls into local extremum. Simulation results...
A new neural network for least squares support vector machines (LS-SVM) learning, which combines LS-SVM with recurrent neural networks, is proposed based on the learning network of standard SVM. It is obtained using Lagrange multipliers directly which eliminates the nonlinear parts of the standard SVM learning network. The proposed network can be used for classification and regression application,...
Starting with principles of neural network and genetic algorithm, new approach, combining genetic algorithm and neural network, of structure optimization were given. Structure optimum target function and design variables set were defined, and with learning algorithm of neural network, non-linear global mapping relationship, between design parameters such as weight, stress, displacement and etc., was...
In this article is presented a method to design neural-genetic optimal controllers that are based on the fusion of a Recurrent Neural Network (RNN) and Genetic Algorithm (GA), these Computational Intelligence (CI) paradigms support the Linear Quadratic (LQR) design. The GA and RNN adaptation proprieties are the great advantage of the proposed approach, because all design is oriented to tune the optimal...
This paper proposes a new global optimization technique in which combines population migration algorithm (PMA) and radial basis function (RBF) neural networks learning algorithm for training RBF neural network. Compared with the traditional RBF training algorithm, the simulation results show that the method has a higher accuracy in a stringency and works well in avoiding sticking in local minima.
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.