Wyniki wyszukiwania - BazTech

1

Backpropagation generalized delta rule for the selective attention Sigma-if artificial neural network

Huk M.

International Journal of Applied Mathematics and Computer Science

|

2012

|

Vol. 22, no. 2

449-459

EN

In this paper the Sigma-if artificial neural network model is considered, which is a generalization of an MLP network with sigmoidal neurons. It was found to be a potentially universal tool for automatic creation of distributed classification and selective attention systems. To overcome the high nonlinearity of the aggregation function of Sigma-if neurons, the training process of the Sigma-if network combines an error backpropagation algorithm with the self-consistency paradigm widely used in physics. But for the same reason, the classical backpropagation delta rule for the MLP network cannot be used. The general equation for the backpropagation generalized delta rule for the Sigma-if neural network is derived and a selection of experimental results that confirm its usefulness are presented.

2

Prawdopodobieństwo błędu klasyfikatorów złożonych dla problemów wieloklasowych

Huk M., Szczepanik M.

Eksploatacja i Niezawodność

|

2011

|

nr 3

12-16

EN

In this paper we consider majority voting of multiple classifiers systems in the case of two-valued decision support for many-class problem. Using an explicit representation of the classification error probability for ensemble binomial voting and two class problem, we obtain general equation for classification error probability for the case under consideration. Thus we are extending theoretical analysis of the given subject initially performed for the two class problem by Hassen and Salamon and still used by Kuncheva and other researchers. This allows us to observe important dependence of maximal posterior error probability of base classifier allowable for building multiple classifiers from the number of considered classes. This indicates the possibility of improving the performance of multiple classifiers for multiclass problems, which may have important implications for their future applications in many fields of science and industry, including the problems of machines diagnostic and systems reliability testing.

PL

W niniejszym artykule rozważamy systemy złożonych klasyfikatorów z głosowaniem większościowym dla przypadku problemów wieloklasowych, wykorzystujące wielowartościowe klasyfikatory bazowe. Stosując bezpośrednią reprezentację prawdopodobieństwa błędnej klasyfikacji dla analogicznych systemów w problemach dwuklasowych, otrzymujemy ogólny wzór na prawdopodobieństwo błędu klasyfikacji w przypadku wieloklasowym. Tym samym rozszerzamy teoretyczne analizy tego zagadnienia pierwotnie przeprowadzone dla problemów dwuklasowych przez Hansena i Salomona i ciagle wykorzystywane przez Kunchevę i innych badaczy. Pozwala nam to zaobserwować istotną zależność maksymalnego dopuszczalnego poziomu prawdopodobieństwa błędów klasyfikatorów bazowych od liczby rozważanych przez nie klas. Wskazuje to na możliwość poprawy parametrów klasyfikatorów złożonych dla problemów wieloklasowych, co może mieć niebagatelne znaczenie dla dalszych ich zastosowań w licznych dziedzinach nauki i przemysłu, z uwzględnieniem zagadnień diagnostyki maszyn oraz badania niezawodności systemów.

3

The sigma-if neural network as a method of dynamic selection of decision subspaces for medical reasoning systems

Huk M.

Journal of Medical Informatics & Technologies

|

2004

|

Vol. 7

KB65--73

EN

To-date research in the area of applied medical artificial intelligence systems suggests that it is necessary to focus further on the characteristic requirements of this research field. One of those requirements is related to the need for effective analysis of multidimensional heterogeneous data sets, which poses particular difficulties when considering AI-suggested solutions. Recent works point to the possibility of extending the activation function of a perception to the time domain, thus significantly enhancing the capabilities of neural networks. This change results in the ability to dynamically tune the size of the decision space under consideration, which stems from continuous adaptation of the interneuron connection architecture to the data being classified. Such adaptation reflects the importance of individual decision attributes for the patterns being classified, as defined by the Sigma-if network during its training phase. These characteristics enable effective employment of such networks in solving classification problems, which emerge in medical sciences. The described approach is also a novel, interesting area of neural network research. This article discusses selected aspects of construction as well as training of Sigma-if networks, based on a sample problem of classifying Arabic numeral images.