Title :
Computer networking topological and parametric representations for functional neural networks
Author :
Niznik, Carol A. ; Hoss, Wayne ; Watts, Charles
Abstract :
The correlated biochemistry and electrical-engineering-based development of the entropy theory is formalized for the parametrization of path information, i.e. storage measures. Similarly, the computer and neural network path delays are developed. Both path parameters are illustrated for the computer networking topological representation of two neural function organs and their interconnection, i.e. the star, the hippocampus computer network representation, and the wheel with a hub, the cerebellum computer network representation. The information measures illustrate the topologically motivated data-compression and connectivity differences. The following computer networking issues are realized in the topological neural network conversion to a functional computer network derivation of path delay and storage parameters: load, real-time requests for various topological structures, concurrency in centralized and distributed networks, deadlock and livelock control, and contention resolution. Category theory is utilized as the mathematical formulation for the definition of simulation in this software theory
Keywords :
computer networks; delays; entropy; neural nets; set theory; topology; category theory; cerebellum; computer network representation; computer networking topological representation; concurrency; connectivity differences; contention resolution; data-compression; deadlock; distributed networks; entropy theory; functional neural networks; hippocampus; information measures; livelock control; neural function organs; neural network path delays; parametric representations; path information; star; storage measures; storage parameters; topological neural network; topological structures;
Conference_Titel :
Neural Networks, 1990., 1990 IJCNN International Joint Conference on
Conference_Location :
San Diego, CA, USA
DOI :
10.1109/IJCNN.1990.137903