List of Publications Competitive and Neuronal Computation

Wolfgang Banzhaf

Older papers (from 1993 back) are represented by abstracts only and are available upon email request
We give titles and links. If you click the underlined words in a title you will see an abstract and source information of the paper. If you click the corresponding filename you will retrieve a copy.







Some Abstracts and Sources

TITLE: On the Dynamics of Competition in a simple Artificial Chemistry

AUTHORS: Wolfgang Banzhaf

SOURCE: Nonlinear Phenomena in Complex Systems, Vol. 5 (2002) pp. 318 - 324

ABSTRACT: We examine a simple system of competing and cooperating entities in terms of the speed of settling their competition. It turns out that the larger the degree of cooperativity among entities the quicker the competition is decided. This result, derived in a simple artificial chemistry system, demonstrates that cooperativity is a decisive element of a world of entities competing for resources. It also hints at the fact that growth of complexity (in terms of increasing cooperativity) is a native tendency of such a world.

FILENAME: npcs.pdf (180 kB)

TITLE: Optical Implementation of a Competitive Network

AUTHORS: W. Banzhaf, E. Lange, M. Oita, J. Ohta, K. Kyuma and T. Nakayama

SOURCE: Frontier Decision Support Concepts, Wiley Series in Sixth Generation Computing Technologies Wiley, New York, 1994, pp. 357 -- 390

EDITORS: V.L. Plantamura, B. Soucek, G. Visaggio



TITLE: A Dynamical Implementation of Self-organizing Maps

AUTHORS: Wolfgang Banzhaf and Manfred Schmutz

SOURCE: Proceedings 1st. Int. Conf. on Applied Synergetics and Synergetic Engineering (ICASSE-94), Erlangen, F.G. B\"obel, T. Wagner (Eds.), Fraunhofer Institut IIE, 1994, pp. 66 --- 73

ABSTRACT: The standard learning algorithm for self-organizing maps (SOM) involves the two steps of a search for the best matching neuron and of an update of its weight vectors in the neighborhood of this neuron. In the dynamical implementation discussed here, a competitive dynamics of laterally coupled neurons with diffusive interaction is used to find the best-matching neuron. The resulting neuronal excitation bubbles are used to drive a Hebbian learning algorithm that is similar to the one Kohonen uses. Convergence of the SOM is achieved here by relating time (or number of training steps) to the strength of the diffusive coupling. A standard application of the SOM is used to demonstrate the feasibility of the approach. 

FILENAME: (158 kB)

TITLE: Competition as an organizational principle for massively parallel computers?

AUTHORS: Wolfgang Banzhaf

SOURCE: Proceedings of the Workshop on, Physics and Computation, Dallas, TX, 1992, IEEE Computer Society Press, Los Alamitos, pp. 229 --- 231

ABSTRACT: We discuss the idea of using competition as a guiding principle for organizing a parallel computer. We argue that competitive interactions are ubiquous in many systems and deserve to be looked at in parallel computing. We outline some relevant questions which have to be answered in this context. 


TITLE: Some Notes on Competition among Cell Assemblies

AUTHORS: Wolfgang Banzhaf and Manfred Schmutz

SOURCE: International Journal on Neural Systems, Volume 2 (1992), pp. 303 --- 313

ABSTRACT: We discuss a family of competitive dynamics useful for pattern recognition purposes. Derived from a physical model of mode competition, they generalize former concepts to include populations of cells working as grandmother cell assemblies. Also the notion of unfair competition is introduced. 

FILENAME: To obtain it, please send me an email!

TITLE: Learning in a competitive network

AUTHORS: Wolfgang Banzhaf and Hermann Haken

SOURCE: Neural Networks, Volume 3 (1990), pp. 423 --- 435

ABSTRACT: We consider the abilities of a recently published neural network model to recognize and classify arbitrary patterns. We introduce a learning scheme based on Hebb's rule which allows the system's neuronal cells to specialize on different patterns during learning. The rule which was originally introduced by Kohonen is appropriately modified and applied to the competitive network under study. A variant of the learning dynamics is then derived from an energy functional characterizing the specialization state of the network. Simulations are presented to demonstrate the specialization process for different pattern distributions. 

FILENAME: To obtain it, please send me an email!

Wolfgang Banzhaf
Last updated: May 31, 2010