TY - GEN
T1 - Preventing Computational Chaos in Asynchronous Neural Networks
AU - Barhen, Jacob
AU - Protopopescu, Vladimir
PY - 2003
Y1 - 2003
N2 - One of the primary advantages of artificial neural networks is their inherent ability to perform massively parallel, nonlinear signal processing. However, the asynchronous dynamics underlying the evolution of such networks may often lead to the emergence of computational chaos, which impedes the efficient retrieval of information usually stored in the system's attractors. In this paper, we discuss the implications of chaos in concurrent asynchronous computation, and provide a methodology that prevents its emergence. Our results are illustrated on a widely used neural network model.
AB - One of the primary advantages of artificial neural networks is their inherent ability to perform massively parallel, nonlinear signal processing. However, the asynchronous dynamics underlying the evolution of such networks may often lead to the emergence of computational chaos, which impedes the efficient retrieval of information usually stored in the system's attractors. In this paper, we discuss the implications of chaos in concurrent asynchronous computation, and provide a methodology that prevents its emergence. Our results are illustrated on a widely used neural network model.
UR - http://www.scopus.com/inward/record.url?scp=1442326754&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:1442326754
SN - 0889863415
SN - 9780889863415
T3 - IASTED International Multi-Conference on Applied Informatics
SP - 1290
EP - 1295
BT - 21st IASTED International Multi-Conference on Applied Informatics
T2 - 21st IASTED International Multi-Conference on Applied Informatics
Y2 - 10 February 2003 through 13 February 2003
ER -