A Rule Chaining Architecture Using a Correlation Matrix Memory
This paper describes an architecture based on superimposed distributed representations and distributed associative memories which is capable of performing rule chaining. The use of a distributed representation allows the system to utilise memory efficiently, and the use of superposition reduces the time complexity of a tree search to O(d), where d is the depth of the tree. Our experimental results show that the architecture is capable of rule chaining effectively, but that further investigation is needed to address capacity considerations.
Keywordsrule chaining correlation matrix memory associative memory distributed representation parallel distributed computation
Unable to display preview. Download preview PDF.
- 1.Kohonen, T.: Correlation Matrix Memories. IEEE Transactions on Computers, 353–359 (1972)Google Scholar
- 3.Austin, J.: Parallel Distributed Computation in Vision. In: IEE Colloquium on Neural Networks for Image Processing Applications, pp. 3/1–3/3 (1992)Google Scholar
- 4.Kustrin, D., Austin, J.: Connectionist Propositional Logic A Simple Correlation Matrix Memory Based Reasoning System. In: Wermter, S., Austin, J., Willshaw, D.J. (eds.) Emergent Neural Computational Architectures Based on Neuroscience. LNCS (LNAI), vol. 2036, pp. 534–546. Springer, Heidelberg (2001)CrossRefGoogle Scholar
- 7.Palm, G.: On the Storage Capacity of Associative Memories. In: Neural Assemblies, an Alternative Approach to Artificial Intelligence, pp. 192–199. Springer, New York (1982)Google Scholar