KENTA KUBOTA 久保田健太

久保田健太 Kenta Kubota

連合学習に関する研究に従事．
IEEE Graduate Student Member．電子情報通信学会会員．映像情報メディア学会会員．
E-mail: kubota [at] lmd [dot] ist [dot] hokudai [dot] ac [dot] jp

経歴

〇Kenta Kubota, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Analysis of model merging methods for continual updating of foundation models in distributed data settings,” Applied Sciences, vol. 15, no. 9:5196, pp.1-18, 2025.

〇Kenta Kubota, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Corrosion Detection through Integration of Multi-resolution Semantic Segmentation,” International Technical Conference on Circuits/Systems, Computers, and Communications (ITC-CSCC), 2025.
〇Kenta Kubota, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Analysis of Model Merging for Open-Vocabulary Models with Parameter Efficient Fine-Tuning Leveraging Distributed Data,” IEEE International Conference on Consumer Electronics-Taiwan (ICCE-TW), 2025.
〇Kenta Kubota, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Gradient-oriented Clustered Federated
Learning with Efficient Knowledge Sharing in non-iid Settings,” IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.1-5, 2025.
〇Kenta Kubota, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Balancing Generalization and
Personalization by Sharing Layers in Clustered Federated Learning,” International Workshop on Advanced Image Technology (IWAIT), vol.13510, pp.112-116, 2025.
〇Kenta Kubota, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “MLLM-Based Automatic Exploration of
Editing Prompt for High Engagement Image Generation,” IEEE Global Conference on Consumer Electronics (GCCE), pp.1165-1166, 2024.

〇久保田健太, 藤後廉, 前田圭介, 小川貴弘, 長谷山美紀, “連合学習におけるモデルマージ導入の有効性に関する評価 – 複数のドメインを対象とした画像分類タスクによる検証 -,” 映像情報メディア学会技術報告, vol.49, no.4, pp.321-325, 2025．
〇久保田健太, 藤後廉, 前田圭介, 小川貴弘, 長谷山美紀, “モデルの学習軌跡を用いたパーソナライズド連合学習におけるクライアントの非階層的クラスタリングの精度向上に関する検討,” 令和6年度電気・情報関係学会北海道支部連合大会, pp.177-178, 2024.