Presentation | 2007-05-21 Unbiased Likelihood Backpropagation Learning Masashi SEKINO, Katsumi NITTA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The error backpropagation is one of the popular methods for training an artificial neural network. When the error backpropagation is used for training an artificial neural network, overfitting occurs in the latter half of training. This paper proposes the unbiased likelihood backpropagation learning which is the gradient discent method with unbiased likelihood (information criterion) as a target function. It is expected that the proposed method has better approximation performance because the method explicitly minimize an estimator of the generalization error. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | unbiased learning / neural network / backpropagation / information criterion / overfitting |
Paper # | NC2007-1 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2007/5/14(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) | Unbiased Likelihood Backpropagation Learning |
Sub Title (in English) | |
Keyword(1) | unbiased learning |
Keyword(2) | neural network |
Keyword(3) | backpropagation |
Keyword(4) | information criterion |
Keyword(5) | overfitting |
1st Author's Name | Masashi SEKINO |
1st Author's Affiliation | Tokyo Institute of Technology() |
2nd Author's Name | Katsumi NITTA |
2nd Author's Affiliation | Tokyo Institute of Technology |
Date | 2007-05-21 |
Paper # | NC2007-1 |
Volume (vol) | vol.107 |
Number (no) | 50 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |