Presentation | 1999/5/20 A Larger Family of Objective Functions to which Hopfield Neural Networks can give Globally Optimal Solutions Yoshinori Uesaka, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | A family of objective functions is discussed for which the conjecture, stating that globally optimal solution (not a local one) of Hopfield neural networks may be obtained by starting from an initial point sufficiently close to the origin, truly holds. To the conjecture, Nishi (1998) has been giving an interesting family of objective functions through the eigenvector analysis. The present paper extends his result to the larger family which suggests that the conjecture really holds for a wide range of objective functions. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Combinatorial optimization / Global optimum / Hopfield neural networks / Dynamical system |
Paper # | NC99-3 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 1999/5/20(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) | A Larger Family of Objective Functions to which Hopfield Neural Networks can give Globally Optimal Solutions |
Sub Title (in English) | |
Keyword(1) | Combinatorial optimization |
Keyword(2) | Global optimum |
Keyword(3) | Hopfield neural networks |
Keyword(4) | Dynamical system |
1st Author's Name | Yoshinori Uesaka |
1st Author's Affiliation | Department of Information Sciences, Faculty of Science and Technology, Science University of Tokyo() |
Date | 1999/5/20 |
Paper # | NC99-3 |
Volume (vol) | vol.99 |
Number (no) | 58 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |