| Presentation | 2013-03-14 Learning Process Visualization of LVQ and Learning Observation at Place where Turning Point Exists Yoshiaki KUROSAWA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | LVQ is one of the basic learning methods in the field of pattern recognition and it can be described by probabilistic descent method. From this aspect , it becomes one of steepest descent method applications. By this concept, the movement of recognition parameters, which are all the reference vector values of LVQ, is observable during learning. This observation technique was applied to the case of a handwritten numerical character recognition experiment. The case reported here was happened to be found in the experiment, and it was that the error rates obtained from almost same learning strength became largely different. As a result of the observation, a turning point was found where the two almost same learning processes were divided and introduced to two different learning processes which had largely different results. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | LVQ / probabilistic descent / steepest descent / learning / OCR / visualization / turning point |
| Paper # | PRMU2012-198 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2013/3/7(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) | Learning Process Visualization of LVQ and Learning Observation at Place where Turning Point Exists |
| Sub Title (in English) | |
| Keyword(1) | LVQ |
| Keyword(2) | probabilistic descent |
| Keyword(3) | steepest descent |
| Keyword(4) | learning |
| Keyword(5) | OCR |
| Keyword(6) | visualization |
| Keyword(7) | turning point |
| 1st Author's Name | Yoshiaki KUROSAWA |
| 1st Author's Affiliation | Toshiba Solutions Corporation Fuchu Engineering Center() |
| Date | 2013-03-14 |
| Paper # | PRMU2012-198 |
| Volume (vol) | vol.112 |
| Number (no) | 495 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |