| Presentation | 1998/6/18 Membership Feedback for Context Propagation Kohei INOUE, Kiichi URAHAMA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A pattern recognition model is presented which incorporates temporal or spatial context informations by feedback of membersihp values outputted from WTA neurons at higher levels into pattern selective neurons at lower levels. An unsupervised learning algorithm is derived from the maximum likelihood estimation of prototypes of clusters. Temporal context is considered at first, and a model is examined where a classification membership at each timepoint is fed back at the next timepoint. This model is shown to partition data according to their temporal proximity rather than their proximity in feature spaces. This model is then applied to position invariant pattern recognition of simple images. Spatial context is considered next, and pattern recognition model similar to the one for temporal context is effective for spatial smoothing and disambiguation of noisy and ambiguous spatial signals. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | context dependence / membersihp feedback / position invariant pattern recognition / unsupervised learning |
| Paper # | PRMU98-32 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 1998/6/18(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) | Membership Feedback for Context Propagation |
| Sub Title (in English) | |
| Keyword(1) | context dependence |
| Keyword(2) | membersihp feedback |
| Keyword(3) | position invariant pattern recognition |
| Keyword(4) | unsupervised learning |
| 1st Author's Name | Kohei INOUE |
| 1st Author's Affiliation | Faculty of Visual Communication Design, Kyushu Institute of Design() |
| 2nd Author's Name | Kiichi URAHAMA |
| 2nd Author's Affiliation | Faculty of Visual Communication Design, Kyushu Institute of Design |
| Date | 1998/6/18 |
| Paper # | PRMU98-32 |
| Volume (vol) | vol.98 |
| Number (no) | 126 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |