Title :
Analytical weight shifting models for self-recovery networks
Author :
Lursinsap, C. ; Chu, Hui ; Kim, Jung-Ho
Author_Institution :
Center for Adv. Comput. Studies, Southwestern Louisiana Univ., Lafayette, LA
Abstract :
Summary form only given, as follows. An efficient self-recovery technique for neural networks called weight shifting and its analytical models have been proposed. The technique was applied to recover a network when some faulty links and neurons occurred during the operation. The proposed model is suitable for VLSI inclusion with the network in terms of quick recovery time and silicon area
Keywords :
fault tolerant computing; neural nets; VLSI; faulty links; faulty neurons; neural net recovery; neural networks; recovery time; self-recovery networks; self-recovery technique; silicon area; weight shifting; Analytical models; Neural networks; Silicon; Very large scale integration;
Conference_Titel :
Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-0164-1
DOI :
10.1109/IJCNN.1991.155511
