The Brain & Neural Networks
Online ISSN : 1883-0455
Print ISSN : 1340-766X
ISSN-L : 1340-766X
Search
Advanced search
OR
Browse
Search
Volume 3, Issue 2
Displaying 1-7 of 7 articles from this issue
Show all abstracts
    • |<
    • <
    • 1
    • >
    • >|
  • The Adaptive Network Architecture for a Time-constrained Environment
    Kosei Demura, Atsushi Nakamura, Yuichiro Anzai
    1996 Volume 3 Issue 2 Pages 43-50
    Published: June 05, 1996
    Released on J-STAGE: January 17, 2011
    DOIhttps://doi.org/10.3902/jnns.3.43
    JOURNAL FREE ACCESS
    Show abstractHide abstract
    This paper proposes an adaptive network architecture called Hybrid SOLAR (Supervised One-shot Learning Algorithm for Real number inputs), which is a hybrid algorithm between a one-shot network construction algorithm and an iterative pruning algorithm. Hybrid SOLAR determines a network structure in two stages. In the first stage, Hybrid SOLAR requires only a single presentation of training examples to construct the network and learning is finished. In the second stage, the network prunes redundant weights to improve the generalization ability. Thus, Hybrid SOLAR retains the advantages of those two algorithms. It needs only a single presentation of the training set for learning and the generalization ability is satisfactory.
    View full abstract
    Download PDF (659K)
  • A Phase Locking Theory of Restoration of Rotated Images by Dynamic Link Architecture
    Takeshi Mito, Koji Kurata
    1996 Volume 3 Issue 2 Pages 51-57
    Published: June 05, 1996
    Released on J-STAGE: January 17, 2011
    DOIhttps://doi.org/10.3902/jnns.3.51
    JOURNAL FREE ACCESS
    Show abstractHide abstract
    Dynamic link architecture is self-organizing map applied to image recognition. Mapping formed from data image to sample image. It tends to link two points which have similar local features, and so, can be distorted to some extent. Similarity of two images is evaluated after transforming data image by the mapping. Thus, dynamic link architecture is tolerant to distortion or rotation of data images. We propose a mathematically tractable model of dynamic link architecture, and reduced it to simple phase dynamics, which clearly shows how the model restores rotated data images. Computer simulations verify the theory.
    View full abstract
    Download PDF (1071K)
  • Statistical Neurodynamics of the Sparsely Encoded Associative Memory
    Takashi Shigemitsu, Masato Okada
    1996 Volume 3 Issue 2 Pages 58-64
    Published: June 05, 1996
    Released on J-STAGE: January 17, 2011
    DOIhttps://doi.org/10.3902/jnns.3.58
    JOURNAL FREE ACCESS
    Show abstractHide abstract
    We analyse the dynamics of sparsely encoded associative memory models by means of the statistical neuro-dynamical method. Now the three methods are taken to control the activity of the neuron, (1) optimally controlled threshold at each step, (2) fixed optimal threshold which gives a maximal storage capacity, and (3) adding the fixed threshold to lateral inhibitions. We study the basin of attraction in each of three cases by the statistical neurodynamics and the computer simulations. And, we use continuous dynamics instead of discrete one to improve dynamical properties.
    View full abstract
    Download PDF (493K)
    • |<
    • <
    • 1
    • >
    • >|
feedback
 Top