| Presentation | 2015-03-20 Training of Random Forests Using Covariate Shift on Parallel Distributed Processing Ryoji WAKAYAMA, Akisato KIMURA, Takayoshi YAMASHITA, Yuji YAMAUCHI, Hironobu FUJIYOSHI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Machine learning with big data improves a classification performance but increases computatinal cost for learning. Parallel distributed processing on multiple processors GPUs is often used to reduce processing time. This paper exploits MapReduce, an efficient framework for parallel distributed processing and proposes a novel method for training Random Forests by using the MapReduce framework. At the Map job stage, each worker trains a Transfer Forest with shared data to enhance classification performance. At the Reduce job stage, a reducer removes unreliable decision trees constructed at the Map stage, in order to reduce the computational cost of testing. The proposed method can retain the classification performance, even though unbalanced training samples are assigned to each worker. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Random forests / Transfer learning / Parallel distributed processing / Machine learning / MapReduce |
| Paper # | BioX2014-73,PRMU2014-193 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2015/3/12(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) | Training of Random Forests Using Covariate Shift on Parallel Distributed Processing |
| Sub Title (in English) | |
| Keyword(1) | Random forests |
| Keyword(2) | Transfer learning |
| Keyword(3) | Parallel distributed processing |
| Keyword(4) | Machine learning |
| Keyword(5) | MapReduce |
| 1st Author's Name | Ryoji WAKAYAMA |
| 1st Author's Affiliation | Chubu University() |
| 2nd Author's Name | Akisato KIMURA |
| 2nd Author's Affiliation | Communication Science Laboratories NTT Corporation |
| 3rd Author's Name | Takayoshi YAMASHITA |
| 3rd Author's Affiliation | Chubu University |
| 4th Author's Name | Yuji YAMAUCHI |
| 4th Author's Affiliation | Chubu University |
| 5th Author's Name | Hironobu FUJIYOSHI |
| 5th Author's Affiliation | Chubu University |
| Date | 2015-03-20 |
| Paper # | BioX2014-73,PRMU2014-193 |
| Volume (vol) | vol.114 |
| Number (no) | 521 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |