Operational fault tolerance of the ADAM neural network system

Neural networks offer a powerful parallel distributed computational system which can be trained to solve many problems. They also appear to be inherently fault tolerant. This is unlike a conventional computing system where fault tolerance is achieved by redundancy, thus increasing its overall complexity. The fault tolerance of the advanced distributed associative memory neural network (ADAM) is investigated, focusing on operational use. The effect on the reliability of recall is examined by injecting faults individually, and also in various combinations since correlations between them will influence their overall effect on the system. Simulations are performed of variously configured ADAM systems, and the subsequent analysis of the results indicates that fault tolerance actually arises in a similar manner to N-Modular Redundancy. An ehancement to the ADAM system shows that these results can be generally applied, and also produces more optimal storage