Pseudo-Orthogonalization of Memory Patterns for Associative Memory
Autor: | Kazuyuki Aihara, Takaki Makino, Makito Oku |
---|---|
Rok vydání: | 2013 |
Předmět: |
Theoretical computer science
Artificial neural network Computer Networks and Communications Computer science Association Learning Content-addressable memory Pattern Recognition Automated Computer Science Applications XNOR gate Biomimetics Memory Artificial Intelligence Subtraction Technique Image Interpretation Computer-Assisted Humans Bidirectional associative memory Algorithm Orthogonalization Algorithms Software |
Zdroj: | IEEE Transactions on Neural Networks and Learning Systems. 24:1877-1887 |
ISSN: | 2162-2388 2162-237X |
Popis: | A new method for improving the storage capacity of associative memory models on a neural network is proposed. The storage capacity of the network increases in proportion to the network size in the case of random patterns, but, in general, the capacity suffers from correlation among memory patterns. Numerous solutions to this problem have been proposed so far, but their high computational cost limits their scalability. In this paper, we propose a novel and simple solution that is locally computable without any iteration. Our method involves XNOR masking of the original memory patterns with random patterns, and the masked patterns and masks are concatenated. The resulting decorrelated patterns allow higher storage capacity at the cost of the pattern length. Furthermore, the increase in the pattern length can be reduced through blockwise masking, which results in a small amount of capacity loss. Movie replay and image recognition are presented as examples to demonstrate the scalability of the proposed method. |
Databáze: | OpenAIRE |
Externí odkaz: |