| Presentation | 2001/10/11 The Fisher Kernel and Beyond Koji TSUDA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The kernel methods such as support vector machines need a kernel function between two objects to be defined. When an object is represented as a vector, various kernels, e.g.the Gaussian and polynomial kernels, are available. Also for discrete data such as sequences or graphs, there have been proposed many ad-hoc approaches. However, the kernel which is applicable to general cases was not known before the Fisher kernel was proposed[11]. The Fisher kernel is derived from probabilistic models and applicable to any data as long as a probabilistic model is defined. In this paper, we will give the intuitive explanation for the Fisher kernel in terms of the leave-one-out maps. Also, since the Fisher kernel is not considered as a suitable method for classification, we introduce a specialized kernel for classification, which is called"the TOP kernel"[16]. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | The Fisher kernal / The TOP kernel / Support vector machines / protein classification |
| Paper # | PRMU2001-108,NC2001-58 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2001/10/11(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 | ENG |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | The Fisher Kernel and Beyond |
| Sub Title (in English) | |
| Keyword(1) | The Fisher kernal |
| Keyword(2) | The TOP kernel |
| Keyword(3) | Support vector machines |
| Keyword(4) | protein classification |
| 1st Author's Name | Koji TSUDA |
| 1st Author's Affiliation | AIST Computational Biology Research Center() |
| Date | 2001/10/11 |
| Paper # | PRMU2001-108,NC2001-58 |
| Volume (vol) | vol.101 |
| Number (no) | 362 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |