Koji Noshiro
Bio
Koji Noshiro is a student in Doctoral Program in Computer Science at University of Tsukuba and a member of Multi-Agent System Research Group. He has been a Research Fellow of Japan Society for the Promotion of Science since 2025. He received his Bachelor's and Master's degrees from that university. His main research interests are distributed constraint optimization (DCOP) and metaheuristics. He is also interested in algorithms for reinforcement learning and distributed systems.
Publication
2024
- Koji Noshiro and Koji Hasebe. Counterexamples and amendments to the termination and optimality of ADOPT-based algorithms, Artificial Intelligence, 329:104083, 2024. doi: 10.1016/j.artint.2024.104083
2023
- Koji Noshiro and Koji Hasebe. Flaws of termination and optimality in ADOPT-based algorithms, 32nd International Joint Conference on Artificial Intelligence (IJCAI-23), 8 pages, August 2023. doi: 10.24963/ijcai.2023/222
2022
- 野城滉司,長谷部浩二.ADOPTに対する敵対的エージェントによる攻撃.日本ソフトウェア科学会第39回大会,2022年8月.(poster presentation in Japanese)
- 野城滉司,長谷部浩二.血小板による止血機構を模倣した複数集団型最適化手法.第36回人工知能学会全国大会,2022年6月.(in Japanese)
2021
- 野城滉司,長谷部浩二.血小板による止血機構に基づいた最適化手法,日本ソフトウェア科学会第38回大会,2021年9月.(in Japanese; received Best Presentation Award)
Materials
- Appendix
- ijcai23_appendix.pdf
- Source Code
- https://github.com/nossie0360/adopt-flaw
- Supplementary Material (Traces of Counterexamples)
- aij2023_supplementary.pdf
- Source Code
- https://github.com/nossie0360/adopt-flaw
Contact
- noshiro[at]mas.cs.tsukuba.ac.jp
Last Updated: 2025-04-15