Presentation | 2020-01-31 [Poster Presentation] Experimental Evaluation of Federated Learning in Real Networks Naoya Yoshida, Takayuki Nishio, Masahiro Morikura, Koji Yamamoto, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Federated Learning (FL) is a decentralized learning mechanism, which enables to train machine learning (ML) models using data of mobile devices while keeping all the data on the devices. In FL, a cloud server updates a model by aggregating multiple models updated by the mobile devices. Therefore, the overall training process can become inefficient when some devices are with poor computational resources or wireless channel conditions. In order to mitigate this problem, FL with Client Selection (FedCS) protocol, an extension of FL, has been proposed which solves a client selection problem with resource constraints. While FL and extensions of it have proposed as described above, the bandwidths and computation capability of each client are evaluated by only simulation. Thus, we implement the FL on real devices and evaluate it experimentally. We implement a server and Linux-based single board computers with wireless LAN connection. We performed non-selection FL protocol and FedCS protocol in the environment and solved a image classification problem. The experimental results show that the FL using real devices can learn the model as well as the simulation. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Federated Learning / Experimental Evaluation / Mobile Networks / Scheduling / Machine Learning |
Paper # | SeMI2019-110 |
Date of Issue | 2020-01-23 (SeMI) |
Conference Information | |
Committee | SeMI |
---|---|
Conference Date | 2020/1/30(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Susumu Ishihara(Shizuoka Univ.) |
Vice Chair | Kazuya Monden(Hitachi) / Koji Yamamoto(Kyoto Univ.) |
Secretary | Kazuya Monden(Kyoto Univ.) / Koji Yamamoto(NTT DOCOMO) |
Assistant | Akira Uchiyama(Osaka Univ.) / Kenji Kanai(Waseda Univ.) / Masafumi Hashimoto(Osaka Univ.) |
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) | [Poster Presentation] Experimental Evaluation of Federated Learning in Real Networks |
Sub Title (in English) | |
Keyword(1) | Federated Learning |
Keyword(2) | Experimental Evaluation |
Keyword(3) | Mobile Networks |
Keyword(4) | Scheduling |
Keyword(5) | Machine Learning |
1st Author's Name | Naoya Yoshida |
1st Author's Affiliation | Kyoto University(Kyoto Univ.) |
2nd Author's Name | Takayuki Nishio |
2nd Author's Affiliation | Kyoto University(Kyoto Univ.) |
3rd Author's Name | Masahiro Morikura |
3rd Author's Affiliation | Kyoto University(Kyoto Univ.) |
4th Author's Name | Koji Yamamoto |
4th Author's Affiliation | Kyoto University(Kyoto Univ.) |
Date | 2020-01-31 |
Paper # | SeMI2019-110 |
Volume (vol) | vol.119 |
Number (no) | SeMI-406 |
Page | pp.pp.49-50(SeMI), |
#Pages | 2 |
Date of Issue | 2020-01-23 (SeMI) |