Presentation 2023-03-17
[Encouragement Talk] Evaluating Impact of Users' sojourn time on Learning Performance in Spot-Based Federated Learning
Shota Ono, Taku Yamazaki, Takumi Miyoshi, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In federated learning (FL), each update of the model generates communication, which increases the traffic on the mobile network. Therefore, we propose a spot-based FL method in which the learning process takes place through device-to-device communication among only users present in a particular spot. Although, this method can learn data suitable for a specific spot and provide the service, the number of users in the spot changes as users move around. In this paper, the impact of the users' sojourn time on the learning was evaluated through implementation experiments. The obtained results show that user's departure from the spot tends to degrade learning performance.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Federated learning / Spot / Location information
Paper # ICM2022-56
Date of Issue 2023-03-09 (ICM)

Conference Information
Committee ICM
Conference Date 2023/3/16(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Okinawa Prefectural Museum and Art Museum
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Yuji Nomura(Fujitsu)
Vice Chair Yu Miyoshi(NTT) / Eiji Takahashi(NEC)
Secretary Yu Miyoshi(NTT) / Eiji Takahashi(Fujitsu)
Assistant Ryo Yamamoto(Univ. of Electro-Comm)

Paper Information
Registration To Technical Committee on Information and Communication Management
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) [Encouragement Talk] Evaluating Impact of Users' sojourn time on Learning Performance in Spot-Based Federated Learning
Sub Title (in English)
Keyword(1) Federated learning
Keyword(2) Spot
Keyword(3) Location information
1st Author's Name Shota Ono
1st Author's Affiliation The University of Tokyo(Univ. of Tokyo)
2nd Author's Name Taku Yamazaki
2nd Author's Affiliation Shibaura Institute of Technology(Shibaura Inst. of Tech.)
3rd Author's Name Takumi Miyoshi
3rd Author's Affiliation Shibaura Institute of Technology(Shibaura Inst. of Tech.)
4th Author's Name Akihito Taya
4th Author's Affiliation The University of Tokyo(Univ. of Tokyo)
5th Author's Name Yuuki Nishiyama
5th Author's Affiliation The University of Tokyo(Univ. of Tokyo)
6th Author's Name Kaoru Sezaki
6th Author's Affiliation The University of Tokyo(Univ. of Tokyo)
Date 2023-03-17
Paper # ICM2022-56
Volume (vol) vol.122
Number (no) ICM-442
Page pp.pp.80-83(ICM),
#Pages 4
Date of Issue 2023-03-09 (ICM)