Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지

Neural Network Cubes (N-Cubes) for Unsupervised learning in Gray-Scale noise

  • Lee, Won-Hee (Department of Electrical & Electronics Engineering Chung-Ang University)
  • Published : 1999.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

We consider a class of auto-associative memories namely N-Cubes (Neural-network Cubes) in which 2-D gray-level images and hidden sinusoidal 1-D wavelets are stored in cubical memories. First we develop a learning procedure based upon the least-squares algorithm, Therefore each 2-D training image is mapped into the associated 1-D waveform in the training phase. Second we show how the recall procedure minimizes errors among the orthogonal basis functions in the hidden layer. As a 2-D images ould be retrieved in the recall phase. Simulation results confirm the efficiency and the noise-free properties of N-Cubes.

Keywords

References

  1. IEEE Trans. Syst. Man & Cybern v.18 Bidirectional Associative Memories B. Kosko
  2. Self-Organization and Associative Memory T. Kohonen
  3. IEEE Trans. Neural Networks v.1 Two coding strategies for bidirectionalassociative memory V. F. Wang;J. B. Curz;J. H. Mulligan Jr.
  4. IEEE Transactions on Neural Networks v.5 no.5 Multi-layer Associative Neural Networks (M.A.N.N.) : Storage Capacity vs. Perfect Recall H. Kang
  5. IEEE Trans. Neural Networks v.4 no.4 A Modified Bidirectional Decoding Strategy Based on the BAM Structure W. J. Wang;D. K. Lee
  6. Neighbor-Layer Updating in MBDS for the Recall of Pure Bipolar Patterns in Gray-Scale Noise v.6 no.6 D. L. Lee;W. J. Wang
  7. selected conference papers in Neural Networks Theory, Technology, and Applications (ed.) Multi-layer Associative Neural Networks (M.A.N.N.) : Storage Capacity vs. Noise-free Recall H. Kang;P. K. Simpson(ed.)