Title :
Introduction to adaptive weight lattice neural networks
Author :
Neville, R.S. ; Luk, P.C.K.
Author_Institution :
Dept. of Electr. & Electron. Eng., Hertfordshire Univ., Hatfield, UK
Abstract :
Research into RAM-based neural networks has now been in progress for approximately two decades. In this paper we introduce a novel way to visualise RAM-based neural networks. We also present an alternative way to visualise the modus operandi of these units. The main reason for this is to provide an insight into their properties and dynamical behaviour, which leads to the development of a new visualisation theory of these units as adaptive weight lattice. The investigation aims to shed light on RAM-based neural networks viewed as adaptive weight lattices nets, in order to give a qualitative insight to the international community
Keywords :
adaptive systems; generalisation (artificial intelligence); neural nets; pattern classification; random-access storage; RAM-based neural networks; adaptive weight lattice networks; dynamical behaviour; generalisation; pattern classification; Adaptive systems; Artificial neural networks; Hypercubes; Lattices; Neural networks; Neurons; Phase change random access memory; Resistors; Resumes; Visualization;
Conference_Titel :
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-4859-1
DOI :
10.1109/IJCNN.1998.685976