About

Seiichi Kawata is a researcher whose work bridges reinforcement learning, control systems, and explainable artificial intelligence. His most recent contribution, "Explainable deep learning for sEMG-based similar gesture recognition: A Shapley-value-based solution" (2024, 13 citations), demonstrates a commitment to making AI systems more transparent and interpretable, particularly in biomedical applications. Earlier, Kawata made foundational advances in reinforcement learning, notably through his work on parallel learning systems. In his 2007 paper (10 citations), he introduced a novel strategy using distinct exploration and exploitation agents combined with a Dyna-Q algorithm, enabling faster convergence to optimal policies. He further extended these ideas in 2008 and developed a teaching method using self-organizing maps for reinforcement learning (2004, 10 citations). Kawata also tackled the challenge of continuous state spaces by employing multiple fuzzy-ART networks (2006, 6 citations), allowing reinforcement learning agents to perform effectively in unknown environments. His earlier work in robust control for linear systems with uncertainties (1992, 3 citations) shows a sustained interest in ensuring stability and reliability. With a career spanning over three decades, Kawata’s research consistently focuses on improving the efficiency, transparency, and robustness of learning and control systems.

Research Focus

Key Achievements

4
H-Index
6
Papers
44
Total Citations
7
Avg Citations/Paper
🏆 Most Cited Paper
Explainable deep learning for sEMG-based similar gesture recognition: A Shapley-value-based solution
13 citations · 2024
📈 Most Prolific Year: 2024 (1 Papers)
🤝 Key Collaborators: 10
🏛 Institutions: Ministry of Education of the People's Republic of China, Advanced Institute of Industrial Technology, Tokyo Metropolitan University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago