Universitätspublikationen
Refine
Year of publication
Document Type
- Article (13809)
- Part of Periodical (3491)
- Doctoral Thesis (3333)
- Contribution to a Periodical (2163)
- Book (2110)
- Working Paper (1894)
- Preprint (1811)
- Review (1064)
- Report (909)
- Conference Proceeding (706)
Language
- English (17641)
- German (14024)
- Portuguese (231)
- Spanish (123)
- Italian (66)
- French (64)
- Multiple languages (64)
- Turkish (12)
- Ukrainian (10)
- slo (7)
Keywords
- Deutschland (132)
- COVID-19 (99)
- inflammation (96)
- Financial Institutions (92)
- ECB (69)
- Capital Markets Union (67)
- SARS-CoV-2 (64)
- Financial Markets (61)
- Adorno (58)
- Banking Regulation (52)
Institute
- Medizin (6701)
- Präsidium (5136)
- Physik (3703)
- Wirtschaftswissenschaften (2305)
- Gesellschaftswissenschaften (2021)
- Biowissenschaften (1775)
- Frankfurt Institute for Advanced Studies (FIAS) (1761)
- Informatik (1483)
- Sustainable Architecture for Finance in Europe (SAFE) (1409)
- Biochemie und Chemie (1400)
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.
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
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.
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.
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.
Clathrates are candidate materials for thermoelectric applications because of a number of unique properties. The clathrate I phases in the Ba-Ni-Ge ternary system allow controlled variation of the charge carrier concentration by adjusting the Ni content. Depending on the Ni content, the physical properties vary from metal-like to insulator-like and show a transition from p-type to n-type conduction. Here we present first results on the characterization of millimeter-sized single crystals grown by the Bridgman technique. Single crystals with a composition of Ba8Ni3.5Ge42.1h0.4 show metallic behavior (dp/dT > 0) albeit with high resistivity at room temperature [p (300 K) = 1 mOhm cm]. The charge carrier concentration at 300 K, as determined from Hall-effect measurements, is 2.3 e-/unit cell. The dimensionless thermoelectric figure of merit estimated at 680 K is ZT ~ 0.2. Keywords Clathrates - thermoelectric material - intermetallic compound - nickel
We suggest a new method to compute the spectrum and wave functions of excited states. We construct a stochastic basis of Bargmann link states, drawn from a physical probability density distribution and compute transition amplitudes between stochastic basis states. From such transition matrix we extract wave functions and the energy spectrum. We apply this method toU(1)2+1 lattice gauge theory. As a test we compute the energy spectrum, wave functions and thermodynamical functions of the electric Hamiltonian and compare it with analytical results. We find excellent agreement. We observe scaling of energies and wave functions in the variable of time. We also present first results on a small lattice for the full Hamiltonian including the magnetic term.
Central elements of the Bologna declaration have been implemented in a huge variety of curricula in humanities, social sciences, natural sciences and engineering sciences at German universities. Overall the results have been nothing less than disastrous. Surprisingly, this seems to be the perfect time for German universities to talk about introducing a curriculum that is fully compatible with the Bologna declaration for medical education as well. However, German medical education does not have problems the Bologna declaration is intended to solve, such as quality, mobility, internationalization and employability. It is already in the Post-Bologna age.
Meeting Abstract : Deutsche Gesellschaft für Chirurgie. 125. Kongress der Deutschen Gesellschaft für Chirurgie. Berlin, 22.-25.04.2008 Einleitung: Ein wesentliches Ziel der modernen Perforatorlappen vom Unterbauch (DIEP-flap) für die Brustrekonstruktion nach Mammaamputation ist die Schonung der Rektusmuskulatur. Der Funktionserhalt der Muskulatur ist abhängig von der Präparationstechnik. In unserer Studie wird die Interaktion zwischen der Muskel- und Nervendurchtrennung und der postoperativen Muskelfunktion untersucht. Material und Methoden: Unser Patientenkollektiv umfasst 42 Patienten. Im Zeitraum von 6/04 bis 06/07 wurden 44 DIEP-Lappen an unserer Klink nach dem gleichen operativen Standard von unterschiedlichen Operateuren zur Brustrekonstruktion transferiert. Die Standards beinhalten die beidseitige Präparation der Perforatorgefäße des Unterbauches, der SIEA-Gefäße, die Auswahl der 2–4 kräftigsten Perforatoren einer Seite und die schonende Präparation der Rektusmuskulatur und der motorischen Nervenäste.In einer prospektiven monozentrischen Studie haben wir die Rektusmuskulatur präoperativ und 6 Monate postoperativ untersucht. Für die Funktionsanalyse wurde sowohl die Myosonografie der Rektusmuskulatur als auch eine klinischen Untersuchung angewandt. Intraoperativ wurde die Anzahl und Lokalisation der Perforatoren, die Länge der gespreizten Muskulatur, die Länge der durchtrennten Muskulatur und die Anzahl und Lokalisation der durchtrennten intramuskulären Nerven in einer Skizze eingetragen. Die Relation zwischen der intraoperativen Muskel- und Nervenschädigung und der postoperativen Funktion wurde analysiert. Ergebnisse: Bei der Hebung des DIEP – flaps wurden im Durchschnitt 10,8 cm Muskulatur gespreizt, 8,2 cm Muskulatur getunnelt und 2,5 cm Muskulatur durchtrennt. In 41% (18 Pat) wurde 1 motorischer Nervenast durchtrennt, in 27,3% (12 Pat) waren es 2 und in 13,6% (6 Pat) 3 Nervenäste. Bei der klinischen Untersuchung 6 Monate postoperativ hatten 8 Patientinnen noch funktionelle Störungen beim Heben schwerer Gegenstände. Myosonografisch fand sich bei 3 Patientinnen eine Funktionsminderung: 1 vollständiger Funktionsverlust der Muskulatur mit Relaxatio, 2 relevante Minderungen der Kontraktilität Bei keiner Patientin fand sich eine Bauchwandhernie. Bei allen Patientinnen mit einer Beeinträchtigung der Muskulatur waren mind. 2 motorische Nervenäste durchtrennt worden. Schlussfolgerung: Die klinische und myosonografische Funktionsanalyse der Bauchwand ermöglicht die Erstellung von Standards zur verbesserten Operationstechnik. Unsere Ergebnisse zeigen, dass die Durchtrennung von 2 oder mehr motorischen Nervenästen vermieden werden muß. Die Länge der durchtrennten und gespreizten Muskulatur ist dagegen von geringerer Bedeutung.