The paper presents a scheme for reducing memory space of a holographic associative memory for content based learning, searching and retrieval of sparse patterns. Holographic associative memory developed on the properties of complex valued Riemann space is one of the most promising models of associative memory. It has demonstrated 10 to 100 times speedup than most other models of associative memories in learning pattern associations with nearly arbitrary level of complexity. The correlation space of the sparse patterns, is also sparse in information, but representationally dense. Therefore, holograph of sparse patterns (such as images) becomes extremely large. In this paper we describe a holographic memory model which projects the sparse holograph on a reduced memory space along all three of its dimensions by unsupervised learning of the stimulus and response patterns. The resulting holographic model also simultaneously increases the encoding, searching and decoding speed.
[1]
Erkki Oja,et al.
Subspace methods of pattern recognition
,
1983
.
[2]
Terence D. Sanger,et al.
Optimal unsupervised learning in a single-layer linear feedforward neural network
,
1989,
Neural Networks.
[3]
David Y. Y. Yun,et al.
Chaotic vectors and a proposal for multidimensional associative network
,
1994,
Electronic Imaging.
[4]
John G. Sutherland.
A Holographic Model of Memory, Learning and Expression
,
1990,
Int. J. Neural Syst..
[5]
Pentti Kanerva,et al.
Sparse Distributed Memory
,
1988
.