Title :
Neural networks trained for associative memory
Author :
Xiaohong, Bao ; Yingmin, Jia
Author_Institution :
Seventh Res. Div., Beijing Univ. of Aeronaut. & Astronaut., China
Abstract :
This paper presents a learning method for designing associative memories using recurrent feedforward neural networks (RFNNs) which can also be implemented by fully interconnected recurrent neural networks (FIRNNs) attached on a linear output layer. This technique guarantees that desired memories are stored and are attractive over the prescribed domains. The learning method can be traced back to the BP algorithm
Keywords :
backpropagation; content-addressable storage; feedforward neural nets; recurrent neural nets; associative memory; fully interconnected recurrent neural networks; learning method; linear output layer; recurrent feedforward neural networks; Associative memory; Bismuth; Design methodology; Differential equations; Feedforward neural networks; Guidelines; Learning systems; Neural networks; Recurrent neural networks;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.488891
