We suggest the distributed associative memory (DAM) model for distributed and fault-tolerant computation as related to retrieval tasks. The fault tolerance is with respect to noise in the input key data and/or local and global failures in the memory itself. We have developed working models for fault-tolerant image recognition and database information retrieval backed up by experimental results which show the feasibility of such an approach.
- Database retrieval
- distributed associative memory
- distributed computation
- fault tolerance
- neural networks