DocumentCode :
2698517
Title :
The `detailed balance´ net: a stable asymmetric artificial neural system for unsupervised learning
Author :
Hollatz, J. ; Schurmann, B.
fYear :
1990
fDate :
17-21 June 1990
Firstpage :
453
Abstract :
A dynamically stable artificial neural network with graded-response neurons employing an unsupervised learning rule for connection weights of restricted asymmetry is investigated. In particular, the quality of performance of the network after fixing the constants entering the model (passive decay constants, forgetting constants, asymmetry factors, and the steepness of the signal function) is discussed. Subsequent to an estimation of the passive decay and forgetting constants, based on the stationary solutions of the differential equations describing the dynamics of the net, and of the asymmetry factors, the constants are quantified further by optimizing the recognition rate in a computer simulation for a specific model problem in the highly nonlinear (high-gain) limit. Working in the high-gain limit is justified from the behavior of the storage capacity of the net as a function of the steepness of the signal function. First results for applications to a real-world problem (work-piece recognition) indicate that the numerical values obtained for the constants are independent of net size
Keywords :
learning systems; neural nets; asymmetry factors; computer simulation; connection weights; differential equations; dynamically stable artificial neural network; forgetting constants; graded-response neurons; high-gain limit; passive decay constants; recognition rate; restricted asymmetry; signal function; stable asymmetric artificial neural system; steepness; unsupervised learning rule; work-piece recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 1990., 1990 IJCNN International Joint Conference on
Conference_Location :
San Diego, CA, USA
Type :
conf
DOI :
10.1109/IJCNN.1990.137882
Filename :
5726840
Link To Document :
بازگشت