Presentation | 1999/12/16 Minimum Squared Error Estimation of Normal Loglikelihood Koji Tsuda, Shotaro Akaho, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In pattern recognition based on density estimation, it is important to estimate the likelihood at unlabeled samples under the assumption that the underlying distribution is normal. The most frequently used method is the plug-in method, where the mean and variance of normal distribution is estimated from training samples. The plug-in method is simple and can indirectly estimate the likelihood, but the accuracy is poor. We propose to estimate the likelihood directly by minimizing the squared error and show that the accuracy is substantially improved. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Pattern recognition / Normal loglikelihood / Minimum squared error / Plug-in estimator |
Paper # | PRMU99-176 |
Date of Issue |
Conference Information | |
Committee | PRMU |
---|---|
Conference Date | 1999/12/16(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 | Pattern Recognition and Media Understanding (PRMU) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Minimum Squared Error Estimation of Normal Loglikelihood |
Sub Title (in English) | |
Keyword(1) | Pattern recognition |
Keyword(2) | Normal loglikelihood |
Keyword(3) | Minimum squared error |
Keyword(4) | Plug-in estimator |
1st Author's Name | Koji Tsuda |
1st Author's Affiliation | Electrotechnical Laboratory() |
2nd Author's Name | Shotaro Akaho |
2nd Author's Affiliation | Electrotechnical Laboratory |
Date | 1999/12/16 |
Paper # | PRMU99-176 |
Volume (vol) | vol.99 |
Number (no) | 514 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |