Year of publication
- Informatik (38) (remove)
- Approximator networks and the principle of optimal information distribution (1991)
- It is well known that artificial neural nets can be used as approximators of any continous functions to any desired degree. Nevertheless, for a given application and a given network architecture the non-trivial task rests to determine the necessary number of neurons and the necessary accuracy (number of bits) per weight for a satisfactory operation. In this paper the problem is treated by an information theoretic approach. The values for the weights and thresholds in the approximator network are determined analytically. Furthermore, the accuracy of the weights and the number of neurons are seen as general system parameters which determine the the maximal output information (i.e. the approximation error) by the absolute amount and the relative distribution of information contained in the network. A new principle of optimal information distribution is proposed and the conditions for the optimal system parameters are derived. For the simple, instructive example of a linear approximation of a non-linear, quadratic function, the principle of optimal information distribution gives the the optimal system parameters, i.e. the number of neurons and the different resolutions of the variables.
- The error-bounded descriptional complexity of approximation networks (1993)
- It is well known that artificial neural nets can be used as approximators of any continuous functions to any desired degree and therefore be used e.g. in high - speed, real-time process control. Nevertheless, for a given application and a given network architecture the non-trivial task remains to determine the necessary number of neurons and the necessary accuracy (number of bits) per weight for a satisfactory operation which are critical issues in VLSI and computer implementations of nontrivial tasks. In this paper the accuracy of the weights and the number of neurons are seen as general system parameters which determine the maximal approximation error by the absolute amount and the relative distribution of information contained in the network. We define as the error-bounded network descriptional complexity the minimal number of bits for a class of approximation networks which show a certain approximation error and achieve the conditions for this goal by the new principle of optimal information distribution. For two examples, a simple linear approximation of a non-linear, quadratic function and a non-linear approximation of the inverse kinematic transformation used in robot manipulator control, the principle of optimal information distribution gives the the optimal number of neurons and the resolutions of the variables, i.e. the minimal amount of storage for the neural net. Keywords: Kolmogorov complexity, e-Entropy, rate-distortion theory, approximation networks, information distribution, weight resolutions, Kohonen mapping, robot control.
- A VLSI-design of the minimum entropy neuron (1994)
- One of the most interesting domains of feedforward networks is the processing of sensor signals. There do exist some networks which extract most of the information by implementing the maximum entropy principle for Gaussian sources. This is done by transforming input patterns to the base of eigenvectors of the input autocorrelation matrix with the biggest eigenvalues. The basic building block of these networks is the linear neuron, learning with the Oja learning rule. Nevertheless, some researchers in pattern recognition theory claim that for pattern recognition and classification clustering transformations are needed which reduce the intra-class entropy. This leads to stable, reliable features and is implemented for Gaussian sources by a linear transformation using the eigenvectors with the smallest eigenvalues. In another paper (Brause 1992) it is shown that the basic building block for such a transformation can be implemented by a linear neuron using an Anti-Hebb rule and restricted weights. This paper shows the analog VLSI design for such a building block, using standard modules of multiplication and addition. The most tedious problem in this VLSI-application is the design of an analog vector normalization circuitry. It can be shown that the standard approaches of weight summation will not give the convergence to the eigenvectors for a proper feature transformation. To avoid this problem, our design differs significantly from the standard approaches by computing the real Euclidean norm. Keywords: minimum entropy, principal component analysis, VLSI, neural networks, surface approximation, cluster transformation, weight normalization circuit.
- Self-organized learning in multi-layer networks (1995)
- We present a framework for the self-organized formation of high level learning by a statistical preprocessing of features. The paper focuses first on the formation of the features in the context of layers of feature processing units as a kind of resource-restricted associative multiresolution learning We clame that such an architecture must reach maturity by basic statistical proportions, optimizing the information processing capabilities of each layer. The final symbolic output is learned by pure association of features of different levels and kind of sensorial input. Finally, we also show that common error-correction learning for motor skills can be accomplished also by non-specific associative learning. Keywords: feedforward network layers, maximal information gain, restricted Hebbian learning, cellular neural nets, evolutionary associative learning
- A spectral multi-resolution image encoding network (1995)
- After a short introduction into traditional image transform coding, multirate systems and multiscale signal coding the paper focuses on the subject of image encoding by a neural network. Taking also noise into account a network model is proposed which not only learns the optimal localized basis functions for the transform but also learns to implement a whitening filter by multi-resolution encoding. A simulation showing the multi-resolution capabilitys concludes the contribution.
- Sensor encoding using lateral inhibited, self-organized cellular neural networks (1996)
- The paper focuses on the division of the sensor field into subsets of sensor events and proposes the linear transformation with the smallest achievable error for reproduction: the transform coding approach using the principal component analysis (PCA). For the implementation of the PCA, this paper introduces a new symmetrical, lateral inhibited neural network model, proposes an objective function for it and deduces the corresponding learning rules. The necessary conditions for the learning rate and the inhibition parameter for balancing the crosscorrelations vs. the autocorrelations are computed. The simulation reveals that an increasing inhibition can speed up the convergence process in the beginning slightly. In the remaining paper, the application of the network in picture encoding is discussed. Here, the use of non-completely connected networks for the self-organized formation of templates in cellular neural networks is shown. It turns out that the self-organizing Kohonen map is just the non-linear, first order approximation of a general self-organizing scheme. Hereby, the classical transform picture coding is changed to a parallel, local model of linear transformation by locally changing sets of self-organized eigenvector projections with overlapping input receptive fields. This approach favors an effective, cheap implementation of sensor encoding directly on the sensor chip. Keywords: Transform coding, Principal component analysis, Lateral inhibited network, Cellular neural network, Kohonen map, Self-organized eigenvector jets.
- Using RBF nets in rubber industry process control (1996)
- This paper describes the use of a radial basis function (RBF) neural network. It approximates the process parameters for the extrusion of a rubber profile used in tyre production. After introducing the problem, we describe the RBF net algorithm and the modeling of the industrial problem. The algorithm shows good results even using only a few training samples. It turns out that the „curse of dimensions“ plays an important role in the model. The paper concludes by a discussion of possible systematic error influences and improvements.
- Implementing the SCAN language by neural networks (1996)
- The real-time encryption of pictures is an important subject for many applications, e.g. television broadcast stations, network security, etc. The paper shows how the previously introduced SCAN encryption method can be easily implemented using binary neural network autoassociative memory.
- Individuelle Prognose bei kritisch kranken Patienten mit septischem Schock durch ein neuronales Netz? (1998)
- Im Zeitraum 1. 11. 1993 bis 30. 3. 1997 wurden 1149 allgemeinchirurgische Intensivpatienten prospektiv erfaßt, von denen 114 die Kriterien des septischen Schocks erfüllten. Die Letalität der Patienten mit einem septischen Schock betrug 47,3%. Nach Training eines neuronalen Netzes mit 91 (von insgesamt n = 114) Patienten ergab die Testung bei den verbleibenden 23 Patienten bei der Berücksichtigung von Parameterveränderungen vom 1. auf den 2. Tag des septischen Schocks folgendes Ergebnis: Alle 10 verstorbenen Patienten wurden korrekt als nicht überlebend vorhergesagt, von den 13 Überlebenden wurden 12 korrekt als überlebend vorhergesagt (Sensitivität 100%; Spezifität 92,3%). Schlüsselwörter: Neuronales Netzwerk , Prognose , Intensivpatient.
- Noise suppressing sensor encoding and neural signal orthonormalization (1998)
- In this paper we regard first the situation where parallel channels are disturbed by noise. With the goal of maximal information conservation we deduce the conditions for a transform which "immunizes" the channels against noise influence before the signals are used in later operations. It shows up that the signals have to be decorrelated and normalized by the filter which corresponds for the case of one channel to the classical result of Shannon. Additional simulations for image encoding and decoding show that this constitutes an efficient approach for noise suppression. Furthermore, by a corresponding objective function we deduce the stochastic and deterministic learning rules for a neural network that implements the data orthonormalization. In comparison with other already existing normalization networks our network shows approximately the same in the stochastic case but, by its generic deduction ensures the convergence and enables the use as independent building block in other contexts, e.g. whitening for independent component analysis. Keywords: information conservation, whitening filter, data orthonormalization network, image encoding, noise suppression.