Presentation | 1994/6/28 On the Number of Hidden Units of a Neural Network Realizing Symmetric Boolean Functions Akira Hayashi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Rumelhart,Hinton and Williams claimed that at least n hidden units were required to solve the n-bit parity problem by a 3-layer neural network with no direct connection between the input and output units.We propose a 3-layer network with just 2 hidden units of which assumed input-output characteristics is different from the usual logistic function. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | neural networks / parity function / hidden units |
Paper # | NC94-23 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 1994/6/28(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 Number of Hidden Units of a Neural Network Realizing Symmetric Boolean Functions |
Sub Title (in English) | |
Keyword(1) | neural networks |
Keyword(2) | parity function |
Keyword(3) | hidden units |
1st Author's Name | Akira Hayashi |
1st Author's Affiliation | Kanazawa Institute of Technology() |
Date | 1994/6/28 |
Paper # | NC94-23 |
Volume (vol) | vol.94 |
Number (no) | 129 |
Page | pp.pp.- |
#Pages | 5 |
Date of Issue |