Terminal attractor based back propagation learning for feedforward neural networks

The convergence of a class of stochastic pseudo-gradient algorithms driven by correlated data sequences is considered in this paper. The obtained results are applied to a learning algorithm for feedforward neural networks and sufficient conditions for its convergence are determined.

A fast and robust algorithm for training feedforward neural networks (FNNs) by using a variable forgetting factor and U-D factorization-based recursive prediction error (RPE) method is proposed. In comparison with the backpropagation (BP) and RPE based learning algorithms, the proposed algorithm, called UD-RPE, can provide much more accurate learning results in fewer iterations with fewer hidden nodes and improve convergence rate and numerical stability (robustness). In addition, it is less sensitive to start-up parameters, such as initial weights and initial covariance matrix, and the randomness in the observed data. It also has good generalization ability and needs less learning time. Simulation results of nonlinear dynamic system modeling and identification show that the algorithm proposed here is an effective and efficient learning algorithm for FNNs.

A new fast learning algorithm for a multilayer feedforward neural network is presented. The new algorithm, based on an innovative variable step size and gradient averaging, has excellent properties to overcome major drawbacks of the backpropagation algorithm and shows high convergence speed. The improved performance of the new algorithm in comparison with other algorithms is verified via computer simulations.

This paper presents the development of several efficient learning algorithms for neural networks (ELEANNE). The ELEANNE 1 and ELEANNE 2 are two recursive least-squares learning algorithms, proposed for training single-layered neural networks with analog output. This paper also proposes a new optimization strategy for training single-layered neural networks, which provides the basis for the development of a variety of efficient learning algorithms. This optimization strategy is the source of the ELEANNE 3, a second-order learning algorithm for training single-layered neural networks with binary output. A simplified version of this algorithm, called ELEANNE 4, is also derived on the basis of some simplifying but reasonable assumptions. The two algorithms developed for single-layered neural networks provide the basis for the derivation of ELEANNE 5 and ELEANNE 6, which are proposed for training multilayered neural networks with binary output. The ELEANNE 7 is an efficient algorithm developed for training multilayered neural networks with either binary or analog output. >

In this paper a general class of fast learning algorithms for feedforward neural networks is introduced and described. The approach exploits the separability of each layer into linear and nonlinear blocks and consists of two steps. The first step is the descent of the error functional in the space of the outputs of the linear blocks (descent in the neuron space), which can be performed using any preferred optimization strategy. In the second step, each linear block is optimized separately by using a least squares (LS) criterion. To demonstrate the effectiveness of the new approach, a detailed treatment of a gradient descent in the neuron space is conducted. The main properties of this approach are the higher speed of convergence with respect to methods that employ an ordinary gradient descent in the weight space backpropagation (BP), better numerical conditioning, and lower computational cost compared to techniques based on the Hessian matrix. The numerical stability is assured by the use of robust LS linear system solvers, operating directly on the input data of each layer. Experimental results obtained in three problems are described, which confirm the effectiveness of the new method.

In order to solve the problem that the traditional long-term high-speed traffic forecasting algorithm is affected by the approximation ability of the function and easy to fall into the local mass value, we wrote a multivariate-based highway traffic forecasting algorithm scaling and convolutional networks. Because the feedforward wavelet neural network algorithm predicts the short-term traffic flow in different areas, it is necessary to examine the ability to predict the difference between different models. From the standard feedforward wavelet neural network algorithm using global optimization capabilities, we improve the wolf pack algorithm, improve the search accuracy of the algorithm, get the best solution of the estimated value of the work according to the search results when completing the research objectives, and get the ability to predict the work of the model. Feedforward neural network algorithm: we develop and obtain the best short-term high-speed traffic forecast values. The results are as follows: after using the author’s algorithm, the processing time increases by 1.5 seconds, but the average percentage of errors decreases by more than 50%, in fact the error and the root mean square error decreased by about 30%, and the smoothing coefficient increased by about 1%. The prediction of the author’s algorithm for short-term high-speed traffic is better than the wavelet neural network prediction algorithm, and the prediction accuracy and stability of the author’s algorithm are higher.

A comparison was carried out of several learning algorithms for training feedforward neural networks with linear threshold units. These learning algorithms do not require an a priori network architecture, but add neurons at will during training. The performance of these algorithms was compared by using training sets with a particular correlation of the input patterns over the full range of possible correlations. For binary input patterns we present a fast method for the selection of input patterns that can be identified by a single neuron. This method is not based on the perceptron learning rule.

This paper presents an efficient learning algorithm for complex-valued feedforward neural networks with application to classification problems. It simplifies complex-valued neural networks learning by using the forward-only computation rather than traditional forward and backward computations. By incorporating the forward-only computation, the complex-valued Levenberg-Marquardt algorithm becomes more efficient. Comparison results of computation cost show that the proposed forward-only complex-valued learning algorithm can be faster than the traditional implementation of the Levenberg-Marquardt algorithm.

Linear programming algorithm for neural networks

Presents an algorithm of dynamic node creation and weights learning for a hybrid feedforward neural network (HFNN) which consists of a linear model and a multilayer neural network. The algorithm is only based on linear least squares, and no iterative learning process is needed. According to the demand of model precision, the algorithm determines the best weights of network and the minimal number of hidden nodes automatically. Compared with the well-known backpropagation network, simulation results show that the new learning algorithm for the HFNN is efficient in model precision, rate of convergence and generalization ability.

New learning algorithms for an adaptive non-linear forward predictor which is based on a Pipelined Recurrent Neural Network (PRNN) are presented. A computationally efficient Gradient Descent (GD) algorithm, as well as a novel Extended Recursive Least Squares (ERLS) algorithm are tested on the predictor. Simulation studies, based on three speech signals, which have been made public and are available on the World Wide Web (WWW), show that the non-linear predictor does not perform satisfactorily when the previously proposed gradient descent algorithm was used. The steepest descent algorithm is shown to yield a poor performance in terms of the prediction error gain, whereas consistently improved results are obtained using the ERLS algorithm. The merit of the non-linear predictor structure is confirmed by yielding approximately 2 dB higher prediction gain than only a linear structure predictor, which uses the conventional Recursive Least Squares (RLS) algorithm.

Davidon least squares-based learning algorithm for feedforward neural networks

Restricted Boltzmann machine to determine the input weights for extreme learning machines

Based on the critical position of the endpoint quality prediction for basic oxygen furnaces (BOFs) in steelmaking, and the latest results in computational intelligence (CI), this paper deals with the development of a novel memetic algorithm (MA) for neural network (NN) learning. Included in this is the integration of extremal optimization (EO) and Levenberg-Marquardt (LM) gradient search, and its application in BOF endpoint quality prediction. The fundamental analysis reveals that the proposed EO-LM algorithm may provide superior performance in generalization, computation efficiency, and avoid local minima, compared to traditional NN learning methods. Experimental results with production-scale BOF data show that the proposed method can effectively improve the NN model for BOF endpoint quality prediction.

Learning algorithms in neural networks: recent results