- [Home]
- [Research achievement]
- [Research achievement detail]
| Title | Classification Algorithms for Generalized Label Noise Model with Unknown Parameter (in Japanese) |
|---|---|
| Authors | Tota Suko 、Goki Yasuda 、Shunsuke Horii 、Manabu Kobayashi |
| Released Year | 2018 |
| Format | Conference |
| Category | Knowledge information processing |
| Jounal Name | |
| Jounal Page | vol. IEICE-118, no.284, pp.361-366 |
| Published Year | 2018 |
| Published Month | 11 |
| Abstract （English） |
In classification problem, there is a case where noise is added to the label. The generalized label noise model is a model that unifiedly expresses various label noises that are handled in semi-supervised learning, learning from positive and unlabeled data and learning from data including the outliers, and so on. We propose a classification algorithm using the EM algorithm for this model with unknown parameters. We evaluate the performance of the proposed algorithm by numerical experiments. |
| Note （English） |
2 |
| Manuscript | |
| Presentation |
