| Presentation | 2022-01-21 [Short Paper] Asynchronous Gradient-Boosted Decision Trees for Distributed Sensing Devices Yui Yamashita, Akihito Taya, Yoshito Tobe, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Recently, wearable devices that install multiple sensors have been widely used. Although sensor data from these devices are used to train machine learning models, to improve performance of the model of machine learning requires significant amount of data. Federated learning (FL) is proposed as a cooperative distributed machine learning without sharing personal data, but it still requires a server to manage devices and aggregate models in a central way. This paper proposes a serverless collaborative learning algorithm to reduce the maintenance cost of the server and prevent only good models are concentrated in a few companies. The proposed method adopts gradient boosting decision tree (GBDT) so that devices with a limited computation resources can join in the collaborative learning. Moreover, the devices can asynchronously share the models in collaborative learning to improving learning speed considering a difference in calculation speed among the devices. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Internet of Things / Distributed learning / Gradient boosting decision tree / Federated learning |
| Paper # | SeMI2021-64 |
| Date of Issue | 2022-01-13 (SeMI) |
| Conference Information | |
| Committee | SeMI |
|---|---|
| Conference Date | 2022/1/20(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Koji Yamamoto(Kyoto Univ.) |
| Vice Chair | Kazuya Monden(Hitachi) / Yasunori Owada(NICT) |
| Secretary | Kazuya Monden(Cyber Univ.) / Yasunori Owada(Waseda Univ.) |
| Assistant | Yuki Katsumata(NTT DOCOMO) / Akihito Taya(Aoyama Gakuin Univ.) / Yu Nakayama(Tokyo Univ. of Agri. and Tech.) |
| Paper Information | |
| Registration To | Technical Committee on Sensor Network and Mobile Intelligence |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | [Short Paper] Asynchronous Gradient-Boosted Decision Trees for Distributed Sensing Devices |
| Sub Title (in English) | |
| Keyword(1) | Internet of Things |
| Keyword(2) | Distributed learning |
| Keyword(3) | Gradient boosting decision tree |
| Keyword(4) | Federated learning |
| 1st Author's Name | Yui Yamashita |
| 1st Author's Affiliation | Aoyama Gakuin University(Aoyama Gakuin Univ.) |
| 2nd Author's Name | Akihito Taya |
| 2nd Author's Affiliation | Aoyama Gakuin University(Aoyama Gakuin Univ.) |
| 3rd Author's Name | Yoshito Tobe |
| 3rd Author's Affiliation | Aoyama Gakuin University(Aoyama Gakuin Univ.) |
| Date | 2022-01-21 |
| Paper # | SeMI2021-64 |
| Volume (vol) | vol.121 |
| Number (no) | SeMI-333 |
| Page | pp.pp.45-47(SeMI), |
| #Pages | 3 |
| Date of Issue | 2022-01-13 (SeMI) |