Title :
Learning and complexity minimization methods of diophantine and splines neural networks with self-organizing architecture
Author_Institution :
Inst. of Inf. & Autom., Acad. of Sci., St. Petersburg
Abstract :
Diophantine neural networks described by polynomial and splines with integer paramaters are considered. Recurrent and non-recurrent learning algorithms and complexity minimization methods and self-organizing architecture of diophantine and splines neural networks are offered
Keywords :
computational complexity; learning (artificial intelligence); multilayer perceptrons; recurrent neural nets; self-organising feature maps; splines (mathematics); complexity minimization methods; diophantine neural networks; nonrecurrent learning algorithms; recurrent learning algorithms; self-organizing architecture; splines neural networks; Equations; Minimization methods; Network synthesis; Neural networks;
Conference_Titel :
Neuroinformatics and Neurocomputers, 1995., Second International Symposium on
Conference_Location :
Rostov on Don
Print_ISBN :
0-7803-2512-5
DOI :
10.1109/ISNINC.1995.480860
