Title :
Vestibulo colic reflex of cats modelled by a fault tolerant neural network
Author :
Rainoldi, Alberto ; Gianoglio, Paolo
Author_Institution :
Torino Univ., Italy
Abstract :
The realization of a dynamic temporally dependent neural model of the vestibulo-colic reflex of the cat is described. Using a modified feedforward backpropagation algorithm, a network was trained to perform the natural behavior of the cat experimentally measured by biologists. The fault tolerance property of the network was verified on the patterns and the physical structure of the cat sensorial devices and on the weight matrix. Use of neural technology to build fault tolerant devices is discussed
Keywords :
backpropagation; biology computing; digital simulation; fault tolerant computing; feedforward neural nets; mechanoception; neurophysiology; cat; dynamic temporally dependent neural model; fault-tolerant neural network; modified feedforward backpropagation algorithm; vestibulo-colic reflex; Animals; Biological system modeling; Biological systems; Cats; Fault tolerance; Fault tolerant systems; Muscles; Neck; Neural networks; Video recording;
Conference_Titel :
Neural Networks, 1992. IJCNN., International Joint Conference on
Conference_Location :
Baltimore, MD
Print_ISBN :
0-7803-0559-0
DOI :
10.1109/IJCNN.1992.227174
