Presentation | 1996/3/18 On the Basins of the Associative Memory Using the Boltzmann Machine Learning Tetsuya Kojima, Tsutomu Date, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The Hopfield neural network should be able to recall the given nominal patterns from the "noisy" input patterns in order to work well as the associative memory. The size of the basins of the nominal patterns characterizes this property well. In this study, we investigate the percentage at which the nominal patterns can be recalled perfectly from the inputs having various direction cosines with the nominal one when the Boltzmann machine learning is used. It is also shown that such percentages about the inputs with less noise increase with the number of units when the memory rate is under some critical value. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | associative memory / Boltzmann machine learning / basin of attraction |
Paper # | NC-95-130 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 1996/3/18(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Neurocomputing (NC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | On the Basins of the Associative Memory Using the Boltzmann Machine Learning |
Sub Title (in English) | |
Keyword(1) | associative memory |
Keyword(2) | Boltzmann machine learning |
Keyword(3) | basin of attraction |
1st Author's Name | Tetsuya Kojima |
1st Author's Affiliation | Graduate School of Engineering, Hokkaido University() |
2nd Author's Name | Tsutomu Date |
2nd Author's Affiliation | Graduate School of Engineering, Hokkaido University |
Date | 1996/3/18 |
Paper # | NC-95-130 |
Volume (vol) | vol.95 |
Number (no) | 598 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |