Rita Lovassy

24725114800

Publications - 23

Learning the optimal parameter of the Hamacher t-norm applied for fuzzy-rule-based model extraction

László Gál Rita Lovassy L. T. Koczy I. J. Rudas

Publication Name: Neural Computing and Applications

Publication Date: 2014-01-01

Volume: 24

Issue: 1

Page Range: 133-142

Description:

Mamdani-type inference systems with trapezoidal-shaped fuzzy membership functions play a crucial role in a wide variety of engineering systems, including real-time control, transportation and logistics, network management, etc. The automatic identification or construction of such fuzzy systems input output data is one of the key problems in modeling. In the past years, the authors have investigated several different fuzzy t-norms, among others, algebraic and trigonometric ones, and the Hamacher product by substituting the standard "min" t-norm operation, in order to achieve better model fitting. In the present paper, the focus is on examining the general parametric Hamacher t-norm, where the free parameter quite essentially influences the quality of modeling and the learning capability of the model identification system. Based on a wide scope of simulation experiments, a quasi-optimal interval for the value of the Hamacher operator is proposed. © 2013 Springer-Verlag London.

Open Access: Yes

DOI: 10.1007/s00521-013-1499-3

Non-parametric and parametric t-norms applied in fuzzy rule extraction

László Gál Rita Lovassy L. T. Koczy

Publication Name: Iccc 2013 IEEE 9th International Conference on Computational Cybernetics Proceedings

Publication Date: 2013-11-07

Volume: Unknown

Issue: Unknown

Page Range: 299-302

Description:

In this paper we propose non-parametric t-norms such as algebraic, trigonometric and Hamacher product, furthermore parametric Hamacher t-norm in Mamdani type inference systems. Various models with trapezoidal shaped fuzzy membership function are applied in order to improve the efficiency of bacterial memetic algorithm in automatic fuzzy rule identification. © 2013 IEEE.

Open Access: Yes

DOI: 10.1109/ICCCyb.2013.6617607

Progressive bacterial algorithm

László Gál Rita Lovassy L. T. Koczy

Publication Name: Cinti 2012 13th IEEE International Symposium on Computational Intelligence and Informatics Proceedings

Publication Date: 2012-12-01

Volume: Unknown

Issue: Unknown

Page Range: 317-322

Description:

The purpose of this paper is to present a new version of the Bacterial Algorithms used for fuzzy rule base extraction called Progressive Bacterial Algorithm. In order to explore high quality models with very good speed of convergence towards the optimal rule base, we develop an improved version of the Bacterial Evolutionary and former Bacterial Memetic Algorithms. It is shown, in case of multidimensional reference problems, by comparing with existing methods, that an efficient and fast convergent tool is obtained. © 2012 IEEE.

Open Access: Yes

DOI: 10.1109/CINTI.2012.6496782

Fuzzy Flip-Flop based Neural Networks as a novel implementation possibility of multilayer perceptrons

László Gál Rita Lovassy Árpád Tóth L. T. Koczy I. J. Rudas

Publication Name: 2012 IEEE I2mtc International Instrumentation and Measurement Technology Conference Proceedings

Publication Date: 2012-07-30

Volume: Unknown

Issue: Unknown

Page Range: 280-285

Description:

Fuzzy Flip-Flop based Neural Networks (FNN) constructed from fuzzy D flip-flops are studied as a novel technique to implement multilayer perceptrons. The starting point of this approach is the concept of fuzzy flip-flop (F 3), as the extension of the binary counterpart. Fuzzy D flip-flop based neurons are viewed, as sigmoid function generators. Their characteristic equations contain simple fuzzy operations, thus enabling easy implementability. FNNs have an interconnected fuzzy neuron structure composed from a large number of neurons acting in parallel which are capable of learning, and are suitable for function approximation. In this paper we propose the FPGA implementation of ukasiewicz operations, furthermore of fuzzy D flip-flop neurons based on Łukasiewicz norms. © 2012 IEEE.

Open Access: Yes

DOI: 10.1109/I2MTC.2012.6229326

Fuzzy neural networks stability in terms of the number of hidden layers

László Gál Rita Lovassy L. T. Koczy I. J. Rudas

Publication Name: 12th IEEE International Symposium on Computational Intelligence and Informatics Cinti 2011 Proceedings

Publication Date: 2011-12-01

Volume: Unknown

Issue: Unknown

Page Range: 323-328

Description:

This paper introduces an approach for studying the stability, and generalization capability of one and two hidden layer Fuzzy Flip-Flop based Neural Networks (FNNs) with various fuzzy operators. By employing fuzzy flip-flop neurons as sigmoid function generators, novel function approximators are established that also avoid overfitting in the case of test data containing noisy items in the form of outliers. It is shown, by comparing with existing standard tansig function based approaches that reducing the network complexity networks with comparable stability are obtained. Finally, examples are given to illustrate the effect of the hidden layer number of neural networks. © 2011 IEEE.

Open Access: Yes

DOI: 10.1109/CINTI.2011.6108523

Generalization capability of neural networks based on fuzzy operators

László Gál Rita Lovassy L. T. Koczy

Publication Name: Applied and Computational Mathematics

Publication Date: 2011-01-01

Volume: 10

Issue: 2

Page Range: 340-355

Description:

This paper discusses the generalization capability of neural networks based on various fuzzy operators introduced earlier by the authors as Fuzzy Flip-Flop based Neural Networks (FNNs), in comparison with standard (e.g. tansig function based, MATLAB Neural Network Toolbox type) networks in the frame of simple function approximation problems. Various fuzzy neurons, one of them based on a pair of new fuzzy intersection and union, and several other selected well known fuzzy operators (£ukasiewicz and Dombi operators) combined with standard negation have been proposed as suitable for the construction of novel FNNs. We briefly present the sigmoid function generators derived from fuzzy J-K and D flip-flops. An advantage of such FNNs is their easy hardware implementability. The experimental results show that these FNNs provide rather good generalization performance, with far better mathematical stability than the standard tansig based neural networks and are more suitable to avoid overfitting in the case of test data containing noisy items in the form of outliers.

Open Access: Yes

DOI: DOI not available

Robustness of fuzzy flip-flop based neural networks

László Gál Rita Lovassy L. T. Koczy

Publication Name: 11th IEEE International Symposium on Computational Intelligence and Informatics Cinti 2010 Proceedings

Publication Date: 2010-12-01

Volume: Unknown

Issue: Unknown

Page Range: 207-211

Description:

In this paper the robustness of three different types of Fuzzy Flip-Flop based Neural Network (FNN) and the standard tansig based neural networks is compared from the various test function approximation goodness points of view. It is tested how well the fuzzy flip-flop based and the simulated neural networks handle the test data sets outlier points. The robust design of the FNN is presented, and the best suitable fuzzy neuron type is emphasized. Furthermore, the sensitivity of fuzzy neural networks to the fuzzy neuron type and hidden layers neuron number is evaluated. ©2010 IEEE.

Open Access: Yes

DOI: 10.1109/CINTI.2010.5672248

Hardware implementation of fuzzy flip-flops based on łukasiewicz norms

Antonio Hernández Zavala László Gál Rita Lovassy Ildar Batyrshin Oscar Camacho Nieto L. T. Koczy

Publication Name: Proceedings of the 9th Wseas International Conference on Applied Computer and Applied Computational Science Acacos 10

Publication Date: 2010-12-01

Volume: Unknown

Issue: Unknown

Page Range: 196-201

Description:

The digital hardware implementation of various fuzzy operations furthermore of fuzzy flip-flops has been the subject of intense study and application. The fuzzy D flip-flop derived from fuzzy J-K one is a single input - single output unit with sigmoid transfer characteristics in some particular cases, proper to use as neuron in a Fuzzy Neural Networks (FNN). In this paper we propose the hardware realization of fuzzy D flip-flops based on Łukasiewicz norms.

Open Access: Yes

DOI: DOI not available

Function approximation performance of Fuzzy Neural Networks based on frequently used fuzzy operations and a pair of new trigonometric norms

László Gál Rita Lovassy L. T. Koczy

Publication Name: 2010 IEEE World Congress on Computational Intelligence Wcci 2010

Publication Date: 2010-11-25

Volume: Unknown

Issue: Unknown

Page Range: Unknown

Description:

A new triangular t-norm and t-conorm are presented. The new fuzzy operations combined with the standard negation are applied in a practical problem, namely, they are proposed as suitable triangular norms for defining a fuzzy flip-flop based neuron. Other fuzzy J-K and D flip-flop based neurons are constructed by using algebraic, Lukasiewicz, Yager, Dombi and Hamacher connectives. The function approximation performance of a Fuzzy Neural Networks (FNN) built up from various fuzzy neurons are evaluated using six increasingly more complicated problems: various sine waves, battery cell charging characteristics, two dimensional trigonometric functions and a six dimensional benchmark problem. It is shown that the new norms lead to FNNs with better approximation properties in some cases than all the previous ones. © 2010 IEEE.

Open Access: Yes

DOI: 10.1109/FUZZY.2010.5584252

Optimization in fuzzy flip-flop neural networks

László Gál Rita Lovassy L. T. Koczy

Publication Name: Studies in Computational Intelligence

Publication Date: 2010-11-03

Volume: 313

Issue: Unknown

Page Range: 337-348

Description:

The fuzzy J-K and D flip-flops present s-shape transfer characteristics in same particular cases. We propose the fuzzy flip-flop neurons; single input-single output units derived from fuzzy flip-flops as sigmoid function generators. The fuzzy neurons-based neural networks, Fuzzy Flip-Flop Neural Networks (FNN) parameters are quasi optimized using a second-order gradient algorithm, the Levenberg-Marquardt method (LM) and an evolutionary algorithm, the Bacterial Memetic Algorithm with Modified Operator Execution Order (BMAM). The quasi optimized FNN's performance based on Dombi and Yager fuzzy operations has been examined with a series of test functions. © 2010 Springer-Verlag Berlin Heidelberg.

Open Access: Yes

DOI: 10.1007/978-3-642-15220-7_27

Three step bacterial memetic algorithm

László Gál Rita Lovassy L. T. Koczy

Publication Name: Ines 2010 14th International Conference on Intelligent Engineering Systems Proceedings

Publication Date: 2010-07-26

Volume: Unknown

Issue: Unknown

Page Range: 31-36

Description:

In order to study the function approximation performance of Fuzzy Neural Networks built up from fuzzy J-K flip-flop neurons a new learning algorithm, the Three Step Bacterial Memetic Algorithm is proposed. Hybrid evolutionary methods that combine genetic type algorithms with "classic" local search have been applied to perform efficient global search. This novel version of the Bacterial Memetic Algorithm with Modified Operator Execution Order (BMAM) is a recently developed technique of hybrid type. This particular merger of evolutionary and gradient based algorithms combining both global and local search consists of bacterial mutation and, as a second step, the Levenberg-Marquardt (LM) method applied for each clone. This LM step saves in this way some potential solutions that could be lost otherwise after each mutation step. As a third step the LM algorithm is recalled for a few iterations for each individual of the population towards reaching the local optimum. In our novel algorithm various kinds of fast algorithm with less complexity, like Quasi-Newton algorithm, Conjugate Gradient algorithm, and two Backpropagation training algorithms: Gradient Descent and Gradient Descent with Adaptive Learning Rate and Momentum are nested in the bacterial mutation. © 2010 IEEE.

Open Access: Yes

DOI: 10.1109/INES.2010.5483817

Parameter optimisation in fuzzy flip-flop-based neural networks

László Gál Rita Lovassy L. ászlóT Kóczy L. T. Koczy

Publication Name: International Journal of Reasoning Based Intelligent Systems

Publication Date: 2010-01-01

Volume: 2

Issue: 3-4

Page Range: 237-243

Description:

This paper presents a method for optimising the parameters of fuzzy flip-flop-based neural networks (FNN) consisting of fuzzy J-K and D flip-flop neurons based on various popular fuzzy operations using bacterial memetic algorithm with the modified operator execution order (BMAM). In early works, the authors proposed the Levenberg-Marquardt algorithm (LM) a widely used second order gradient type training algorithm for fuzzy neural networks variables optimisation. The BMAM local and global search evolutionary approach is a bacterial type memetic algorithm which executes several LM cycles during the bacterial mutation after each mutational step, using the LM method more efficiently. Numerical experiments were performed to show the function approximation capability of various quasi optimised FNN types based on fuzzy J-K and D flip-flop neurons using algebraic, Lukasiewicz, Yager, Dombi, Hamacher and Frank norms, trained with LM method and BMAM algorithm. © 2010 Inderscience Enterprises Ltd.

Open Access: Yes

DOI: 10.1504/IJRIS.2010.036869

Multilayer perceptrons constructed of fuzzy flip-flops

László Gál Rita Lovassy L. T. Koczy

Publication Name: Isciii 09 4th International Symposium on Computational Intelligence and Intelligent Informatics Proceedings

Publication Date: 2009-12-28

Volume: Unknown

Issue: Unknown

Page Range: 9-14

Description:

The target of this paper is to propose a hybrid combination of the three main branches of Computational Intelligence, namely Fuzzy Systems, Neural Networks and Evolutionary Computing. The function approximation properties of fuzzy J-K and D flip-flops based feedforward neural network optimized and trained with a novel evolutionary algorithm based technique; the Bacterial Memetic Algorithm with Modified Operator Execution Order (BMAM) is studied. © 2009 IEEE.

Open Access: Yes

DOI: 10.1109/ISCIII.2009.5342288

Quasi optimization of fuzzy neural networks

László Gál Rita Lovassy L. T. Koczy

Publication Name: 10th International Symposium of Hungarian Researchers on Computational Intelligence and Informatics Cinti 2009

Publication Date: 2009-12-01

Volume: Unknown

Issue: Unknown

Page Range: 303-314

Description:

The fuzzy flip-flop based multilayer perceptron, named Fuzzy Neural Network, FNN is proposed for function approximation. In recent years much effort has been made for the development of a special kind of bacterial memetic algorithm for optimization and training of the fuzzy neural network parameters. In this approach the FNN parameters have been encoded in a chromosome and participate in the bacterial mutation cycle. The quasi optimized FNN's performance based on various fuzzy flip-flop types has been examined with a series of multidimensional input functions.

Open Access: Yes

DOI: DOI not available

Optimizing fuzzy flip-flop based neural networks by bacterial memetic algorithm

László Gál Rita Lovassy L. T. Koczy

Publication Name: 2009 International Fuzzy Systems Association World Congress and 2009 European Society for Fuzzy Logic and Technology Conference Ifsa Eusflat 2009 Proceedings

Publication Date: 2009-12-01

Volume: Unknown

Issue: Unknown

Page Range: 1508-1513

Description:

In our previous work we proposed a Multilayer Perceptron Neural Networks (MLP NN) consisting of fuzzy flipflops (F3) based on various operations. We showed that such kind of fuzzy-neural network had good learning properties. In this paper we propose an evolutionary approach for optimizing fuzzy flip-flop networks (FNN). Various popular fuzzy operation and three different fuzzy flip-flop types will be compared from the point of view of the respective fuzzy-neural networks' approximation capability.

Open Access: Yes

DOI: DOI not available

Function approximation capability of a novel fuzzy flip-flop based Neural Network

László Gál Rita Lovassy L. T. Koczy

Publication Name: Proceedings of the International Joint Conference on Neural Networks

Publication Date: 2009-11-18

Volume: Unknown

Issue: Unknown

Page Range: 1900-1907

Description:

The function approximation capability of various connectionist systems has been one of the most interesting problems. A method for constructing Multilayer Perceptron Neural Networks (MLP NN) with the aid of fuzzy operations based flip-flops able to approximate single and multiple variable functions is proposed. This paper introduces the concept of fuzzy flip-flop based neural network, particularly by deploying three types of fuzzy flip-flops as neurons. A comparative study of feedbacked fuzzy J-K and two kinds of fuzzy D flip-flops used as neurons, based on fuzzy algebraic, Yager, Dombi, Hamacher and Frank operations is given. Simulation results are presented for several test functions. © 2009 IEEE.

Open Access: Yes

DOI: 10.1109/IJCNN.2009.5178849

Applicability of fuzzy flip-flops in the implementation of neural networks

László Gál Rita Lovassy L. T. Koczy

Publication Name: 9th International Symposium of Hungarian Researchers on Computational Intelligence and Informatics Cinti 2008

Publication Date: 2008-12-01

Volume: Unknown

Issue: Unknown

Page Range: 333-344

Description:

The concept of various type fuzzy flip-flops (F3) has already been proposed. We have done some investigations on a large scope of F 3s based on different t-norms and conorms. Also we have shown that a few F3 types are suitable for realizing neurons in multilayer perceptrons. The aim of this paper is to present a comparison of the performance of several type neural networks based on fuzzy J-K and also fuzzy D flip-flops (the latter derived from the former type). The behavior of algebraic, Yager, Dombi and Hamacher type fuzzy flip-flop neural networks are presented. The best fitting t-norm and corresponding fuzzy flip-flop type will be presented in terms of function approximation capability.

Open Access: Yes

DOI: DOI not available

Multilayer perceptron implemented by fuzzy flip-flops

László Gál Rita Lovassy L. T. Koczy

Publication Name: IEEE International Conference on Fuzzy Systems

Publication Date: 2008-11-07

Volume: Unknown

Issue: Unknown

Page Range: 1683-1688

Description:

The paper introduces a novel method for constructing Multilayer Perceptron (MLP) Neural Networks (NN) with the aid of fuzzy systems, particularly by deploying fuzzy J-K flip-flops as neurons. The next state Q(t+1) of the J-K fuzzy flip-flops (F3) in terms of input J can be characterized by a more or less S-shaped function, for each F3 derived from the Yager, Dombi, and Fodor norms and co-norms. In this approach, J represents the neuron input. The other input K is wired to the complemental output (K=1-Q), thus an elementary fuzzy sequential unit with a single input and a single output is received. The algebraic F3 having linear J-Q(t+1) characteristics is added to the above three. The paper proposes the investigation of the possibility of constructing multilayer perceptrons from such real fuzzy hardware units. Each of the four candidates for F3-based neurons is examined for its training capability by evaluating and comparing the approximation capabilities for two different transcendental functions. Simulation results are presented. © 2008 IEEE.

Open Access: Yes

DOI: 10.1109/FUZZY.2008.4630597

Fuzzy flip-flop based neural network as a function approximator

László Gál Rita Lovassy L. T. Koczy

Publication Name: Cimsa 2008 IEEE Conference on Computational Intelligence for Measurement Systems and Applications Proceedings

Publication Date: 2008-09-26

Volume: Unknown

Issue: Unknown

Page Range: 44-49

Description:

Artificial neural networks and fuzzy logic systems, in the context of approximate reasoning, share common features and techniques. A family of fuzzy flip-flops is proposed, based on an artificial neural network-like structure which is suitable for approximating many-input one-output nonlinear functions. The neurons in the multilayer perceptron networks typically employ sigmoidal activation functions. The next state of the fuzzy J-K flip-flops (F3) using Yager and Dombi operators present quasi-S-shaped characteristics. The paper proposes the investigation of the possibility of constructing multilayer perceptrons from such fuzzy units. Each of the two candidates for F 3-based neurons is examined for its training capability by evaluating and comparing the approximation properties in the context of different transcendental functions with one-input and multi-input cases. Simulation results are presented. ©2008 IEEE.

Open Access: Yes

DOI: 10.1109/CIMSA.2008.4595830

S-shaped fuzzy flip-flops

Rita Lovassy L. T. Koczy

Publication Name: 8th International Symposium of Hungarian Researchers on Computational Intelligence and Informatics Cinti 2007

Publication Date: 2007-12-01

Volume: Unknown

Issue: Unknown

Page Range: 383-391

Description:

The multilayer perceptron is an artificial neural network that learns nonlinear function mappings. Nonlinear functions can be represented by multilayer perceptrons with units that use nonlinear activation functions. The neurons in the multilayer perceptron networks typically employ sigmoidal activation function. The next state of the J-K fuzzy flipflops (F3) using Fodor, Yager and Dombi operators present S-shaped characteristics. An interesting aspect of F3-s might be that they have a certain convergent behavior when one of their inputs (e.g. J) is exited repeatedly. If J is considered the equivalent of the traditional input of a neuron (with an adder unit applied before J), K might play a secondary modifier's role, or can just be set fix. The paper proposes the investigation of such possible F3-networks as new alternative types of neural networks.

Open Access: Yes

DOI: DOI not available

Fuzzy flip-flops revisited

Rita Lovassy L. T. Koczy

Publication Name: Advances in Soft Computing

Publication Date: 2007-12-01

Volume: 42

Issue: Unknown

Page Range: 643-652

Description:

J-K flip-flops are elementary digital units providing sequential features/memory functions. Their definitive equation is used both in the minimal disjunctive and conjunctive forms. Fuzzy connectives do not satisfy all Boolean axioms, thus the fuzzy equivalents of these equations result in two non-equivalent definitions, "reset and set type" fuzzy flip-flops (F3) by Hirota & al. when introducing the concept of F3. There are many alternatives for "fuzzifying" digital flip-flops, using standard, algebraic or other connectives. The paper gives an overview of some of the most famous F3-s by presenting their definitions and presenting graphs of the inner state for a typical state value situation. Then a pair of non-associative operators is introduced, and the properties of the respective F3 are discussed. The investigation of possible fuzzy flip-flops is continued by examining Türksen's IVFS, its midpoint values, and by introducing "minimized IVFS" (MIVFS), along with the MIVFS midpoints. © 2007 Springer-Verlag Berlin Heidelberg.

Open Access: Yes

DOI: 10.1007/978-3-540-72434-6_65

Fuzzy flip-flops and neural nets?

Rita Lovassy L. T. Koczy

Publication Name: IEEE International Conference on Fuzzy Systems

Publication Date: 2007-12-01

Volume: Unknown

Issue: Unknown

Page Range: Unknown

Description:

J-K flip-flops are elementary digital units providing sequential features/memory functions. Their definitive equation is used both in the minimal disjunctive and conjunctive forms. Fuzzy connectives do not satisfy all Boolean axioms, thus the fuzzy equivalents of these equations result in two non-equivalent definitions, "reset and set type" fuzzy flip-flops (F3) by Hirota & al. when introducing the concept of F 3. The paper gives an overview of some of the most famous F 3-s by presenting their definitions and graphs. An interesting aspect of F3-s might be that they have a certain convergent behavior when one of their inputs (e.g. J) is exited repeatedly. This is true even if the other input (K) is kept at a constant value. The behavior is more versatile if both inputs are given a series of changing values. If J is considered the equivalent of the traditional input of a neuron (with an adder unit applied before J), K might play a secondary modifier's role, or can just be set fix. The paper encourages to the investigation of such possible F3-networks as new alternative types of neural networks. © 2007 IEEE.

Open Access: Yes

DOI: 10.1109/FUZZY.2007.4295435

Non-associative fuzzy flip-flop with dual set-reset feature

Rita Lovassy L. T. Koczy

Publication Name: Sisy 2006 4th Serbian Hungarian Joint Symposium on Intelligent Systems

Publication Date: 2006-01-01

Volume: Unknown

Issue: Unknown

Page Range: 289-299

Description:

J-K flip-flops are the most general elementary units in sequential digital circuits. By extending Boolean operations to their respective fuzzy counterparts various fuzzy flipflops (F3) can be defined. Because of the axiomatic properties of fuzzy operations are considerably weaker then the properties satisfied by Boolean lattices, the minterm and maxterm type definitions of the same F3 are as a rule not equivalent. In former work we found a unique exception where simulation investigations lead to identical results for all parameter combinations examined. The base of this unique F3 is a pair of non-associative fuzzy connectives. In this paper the exact proof is given for the identity of the two definitions, i.e., for the uniqueness of the definition of this special non-associative F3.

Open Access: Yes

DOI: DOI not available