Title :
Typed expressions evolution of artificial nervous systems
Author_Institution :
Lehrstuhl fur Programmiersprachen, Erlangen-Nurnberg Univ., Germany
Abstract :
We introduce an evolutionary algorithm paradigm for hierarchical, typed expressions. Some common genetic operators on the generated expressions are demonstrated. The evolved structures serve as genotypical codings for artificial organisms. A few examples are given showing how the expression evolution has been used for evolving artificial nervous systems by adaptation of connection structures and neuron functionalities
Keywords :
genetic algorithms; neural nets; neurophysiology; optimisation; adaptation; artificial nervous systems; artificial organisms; connection structures; evolutionary algorithm; evolving artificial nervous systems; expression evolution; genetic operators; genotypical codings; hierarchical typed expressions; neuron functionalities; typed expressions evolution; Artificial neural networks; Biological neural networks; Computer languages; Computer science; Evolutionary computation; Genetics; Jacobian matrices; Nervous system; Neurons; Organisms;
Conference_Titel :
Evolutionary Computation, 1994. IEEE World Congress on Computational Intelligence., Proceedings of the First IEEE Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1899-4
DOI :
10.1109/ICEC.1994.349899
