Yevgeniy Bodyanskiy

Neural network approach to forecasting of quasiperiodic financial time series

A novel neural network approach to forecasting of financial time series based on the presentation of the series as a combination of quasiperiodic components is presented. Separate components may have aliquant, and possibly non-stationary frequencies. All their parameters are estimated in real time in an ensemble of predictors, whose outputs are then optimally combined to obtain the final forecast. Special architecture of artificial neural network and learning algorithms implementing this approach are developed.

An Adaptive Learning Algorithm for a Neuro-fuzzy Network

The paper addresses the problem of online adaptive learning in a neuro-fuzzy network based on Sugeno-type fuzzy inference. A new learning algorithm for tuning of both antecedent and consequent parts of fuzzy rules is proposed. The algorithm is derived from the well-known Marquardt procedure and uses approximation of the Hessian matrix. A characteristic feature of the proposed algorithm is that it does not require time-consuming matrix operations. Simulation results illustrate apcpaltiion to adaptive identification of a nonlinear plant and nonlinear time series prediction.

A hybrid cascade neural network with an optimized pool in each cascade

This paper proposes a new architecture and learning algorithms for a hybrid cascade neural network with pool optimization in each cascade. The proposed system is different from existing cascade systems in its capability to operate in an online mode, which allows it to work with non-stationary and stochastic nonlinear chaotic signals with the required accuracy. Compared to conventional analogs, the proposed system provides computational simplicity and possesses both tracking and filtering capabilities.

Fuzzy Clustering Data Given on the Ordinal Scale Based on Membership and Likelihood Functions Sharing

A task of clustering data given in the ordinal scale under conditions of overlapping clusters has been considered. Its proposed to use an approach based on memberhsip and likelihood functions sharing. A number of performed experiments proved effectiveness of the proposed method. The proposed method is characterized by robustness to outliers due to a way of ordering values while constructing membership functions.

Neuro-Fuzzy kolmogorov's network for time series prediction and pattern classification

In the paper, a novel Neuro-Fuzzy Kolmogorovs Network (NFKN) is considered. The NFKN is based on and is the development of the previously proposed neural and fuzzy systems using the famous Kolmogorovs superposition theorem (KST). The network consists of two layers of neo-fuzzy neurons (NFNs) and is linear in both the hidden and output layer parameters, so it can be trained with very fast and simple procedures: the gradient-descent based learning rule for the hidden layer, and the recursive least squares algorithm for the output layer. The validity of theoretical results and the advantages of the NFKN are confirmed by experiments.

Universal Approximator Employing Neo-Fuzzy Neurons

A novel fuzzy neural network, called Fuzzy Kolmogorov’s Network (FKN), is considered. The network consists of two layers of neo-fuzzy neurons (NFNs) and is linear in both the hidden and output layer parameters, so it can be trained with very fast and computationally efficient procedures. Two-level structure of the rule base helps the FKN avoid the combinatorial explosion in the number of rules, while the antecedent fuzzy sets completely cover the input hyperbox. The number of rules in the FKN depends linearly on the dimensionality of input space. The validity of theoretical results and the advantages of the FKN are confirmed by a comparison with other techniques in benchmark problems and a real-world problem of electrical load forecasting.

Fuzzy Kolmogorov's Network.

A novel fuzzy neural network, called Fuzzy Kolmogorov’s Network (FKN), is proposed. The network consists of two layers of neo-fuzzy neurons (NFNs) and is linear in both the hidden and output layer parameters, so it can be trained with very fast and computationally efficient procedures. The validity of theoretical results and the advantages of the FKN in comparison with other techniques are confirmed by experiments.

Fuzzy Clustering Data Given in the Ordinal Scale

A fuzzy clustering algorithm for multidimensional data is proposed in this article. The data is described by vectors whose components are linguistic variables defined in an ordinal scale. The obtained results confirm the efficiency of the proposed approach.

Analog-Digital Self-Learning Fuzzy Spiking Neural Network in Image Processing Problems

Computational intelligence provides a variety of means that can perform complex image processing in a rather effective way. Among them, self-learning systems, especially selflearning artificial neural networks (self-organizing maps, ART neural networks, ‘BrainState-in-a-Box’ neuromodels, etc.) (Haykin, 1999) and fuzzy clustering systems (fuzzy cmeans, algorithms of Gustafson-Kessel, Yager-Filev, Klawonn-Hoeppner, etc) (Bezdek et al., 2005; Sato-Ilic & Jain, 2006), occupy a significant place as they make it possible to solve a data processing problem in the absence of a priori knowledge of it. While there are many artificial neural networks that can be successfully used in image processing tasks, the most prominent of them are networks of a new, the third generation, commonly known as spiking neural networks (Maass & Bishop, 1998; Gerstner & Kistler, 2002). On the one hand, spiking neural networks are biologically more plausible than neural networks of the previous generations that is of fundamental importance for computational intelligence from theoretical point of view. On the other hand, networks of spiking neurons appeared to be computationally more powerful than conventional neural networks (Maass, 1997b). In addition, complex data processing via artificial neural networks of the second generation is time consuming due to multi-epoch learning; instead, spiking neural networks can perform the same processing tasks much faster as they require a few learning epochs only (Bohte et al., 2002; Berredo, 2005; Meftah et al., 2008; Lindblad & Kinser, 2005). All these facts are causing considerable interest in networks of spiking neurons as a powerful computational intelligence tool for image processing Although spiking neural networks are becoming a popular computational intelligence tool for various technical problems solving, their architecture and functioning are treated in terms of neurophysiology rather than in terms of any technical sciences apparatus in the most research works on engineering subjects. Yet none technically plausible description of spiking neurons functioning has been provided. In contrast to artificial neural networks, fuzzy logic systems are capable of performing accurate and efficient data processing under a priori and current uncertainty, particularly if classes to be separated overlap one another. Integrating artificial neural networks and fuzzy systems together allows of combining capabilities of both in a synergetic way (Jang et al., 1997), thus producing hybrid intelligent systems that achieve high performance and reliability in real life problems solving, particularly in image processing. Obviously,

Outlier Resistant Recursive Fuzzy Clustering Algorithms

The problem of fuzzy clustering on the basis of the probabilistic and possibilistic approaches under the presence of outliers in data is considered. Robust recursive fuzzy clustering algorithms are proposed, which optimize the objective function suitable for clustering data with heavy-tailed distribution density. Advantages of the proposed algorithms in comparison with the well-known fuzzy c-means algorithm are demonstrated in an experiment in clustering and classification of data with outliers. The robustness property results in finding correct cluster prototypes whose locations are not affected by anomalous observations, and in achieving thus higher classification accuracy.