Current Research Problems (2025–2026)
"Where there's a will, there's a way."
- Data Analysis and Knowledge Mining — Time series data analysis and forecasting; text data mining and topic modelling from social media and news sources.
- Multi-Agent Systems Based on LLMs and Diffusion Models — Building coordinated agent pipelines using large language models, revisiting and extending my late-1990s multi-agent coordination research with modern deep learning techniques.
Research Interests
Artificial Intelligence
- Knowledge representation (ontologies, semantic web, topic maps)
- Reasoning: search, constraint satisfaction, probabilistic and Bayesian reasoning
- Machine learning: deep neural networks, transformers, BERT, attention mechanisms
- Transfer learning and model fine-tuning
- Reinforcement learning and autonomous agents (AlphaZero / MuZero paradigm)
Multi-Agent Systems
- LLM-based agent coordination and cooperation mechanisms
- Hierarchical and emergent multi-agent control
- Negotiation, competition, and collective intelligence
Data Science & NLP
- Topic modelling and social media comment analysis
- Time series forecasting (seasonality, deep models)
- Sentiment analysis and information filtering
- Question-answering systems
- Knowledge discovery from scientific and financial corpora
Laboratory Members (2025–2026)
Current Students
| Year | Name |
|---|---|
| B4 | Shinohe Kirito (篠辺暉仁), Nakano Daiki (中野大樹) |
| B3 | Itoh Seina (伊藤聖菜), Kudo Kazuki (工藤和輝), Miyakoshi Eiichi (宮腰永一) |
Alumni (OB/OG)
| Graduation | Name |
|---|---|
| 2025 | Kawamura Naoki (川村直希), Suzuki Shuma (鈴木朱馬) |
| 2024 | Eguro Yuta (江黒勇太), Sugo Keisuke (須郷圭輔) |
| 2023 | Takechi Shiun (竹知詩雲), Saitoh Rina (齋藤里菜), Seto Yu (瀬戸悠) |
| 2022 | Ikuma Hideki (生熊英樹), Abe Touma (阿部冬真) |
News & Updates
- 2026 — Welcome to the new B3 members: Itoh Seina, Kudo Kazuki, Miyakoshi Eiichi.
- 2025 — Congratulations to OB 2025: Kawamura Naoki and Suzuki Shuma.
- 2024 — Paper on LLM agents for social media comment analysis presented at adNLP 2024, Vancouver.
- 2022 — Paper on deep learning comment filtering presented at IEEE WI-IAT 2022, Niagara Falls.