Personal tools
 
You are here: Home → SLMC Home → People → Takeshi Mori

Takeshi Mori

Takeshi Mori pic

 

Takeshi Mori joined the European STIFF Project as a postdoctoral research fellow at the University of Edinburgh in April 2010. He received his PhD in March 2007 from Nara Institute of Science and Technology (NAIST). He was a postdoctoral researcher at Kyoto University from 2008 to 2010, and also at NAIST from 2007 to 2008. He was granted Grant-in-Aid for Young Scientist B, Japan, from 2008 to 2010, and Grant-in-Aid for JSPS Fellows, Japan, from 2005 to 2007. His research interest up to date is in developing artificial intelligence that can automatically adapt to the unknown world. To this end, he has studied to develop several effective algorithms in reinforcement learning based on statistical learning framework, and applying them to difficult problems such as robotics. He receives his M.Eng from NAIST in 2004, and B. Eng from Waseda University in 2002. He was born in Kobe, Japan in 1979.

 

Contact

IPAB, School of Informatics, University of Edinburgh
Informatics Forum, 1.25
10 Crichton Street
Edinburgh EH8 9AB

Email: takeshi.mori [ at ] ed.ac.uk
Tel: +44 (0) 131 6517 195

Research Projects

  • STIFF
    The goal of the project is to equip a highly biomimetic robot hand-arm system with the agility, robustness and versatility that are hallmarks of the human motor system by understanding and mimicking the variable stiffness paradigms that are so effectively employed by the human CNS. See the STIFF project homepage for more details.

Publications

Journal Papers
  • Takeshi Mori and Shin Ishii, Incremental state aggregation for value function estimation in reinforcement learning, IEEE Transactions on Systems, Man and Cybernetics - Part B: Cybernetics, 2011, 41(5), pp.1407-1416
  • Yutaka Nakamura, Takeshi Mori, Masa-aki Sato and Shin Ishii, Reinforcement learning for a biped robot based on a CPG-actor-critic method, Neural Networks, 2007, 20(6), pp.723-735
  • Takeshi Mori, Yutaka Nakamura and Shin Ishii, Efficient sample reuse by natural actor-critic learning based on importance sampling, The IEICE transactions on information and systems (Japanese edition), 2006, J89-D(5), pp.954-966
  • Yutaka Nakamura, Takeshi Mori, Yoichi Tokita, Tomohiro Shibata and Shin Ishii, Off-policy natural policy gradient method for a biped walking using a CPG controller, Journal of robotics and mechatronics, 2005, 17(6), pp.636-644
  • Takeshi Mori, Yutaka Nakamura and Shin Ishii, Reinforcement learning based on a policy gradient method for a biped locomotion, The IEICE transactions on information and systems (Japanese edition), 2005, J88-D-II(6), pp.1080-1089
Conference Papers
  • Takeshi Mori, Matthew Howard and Sethu Vijayakumar, Model-free apprenticeship learning for transfer of human impedance behaviour, 11th IEEE-RAS International Conference on Humanoid Robots (Humanoids), Bled, Slovenia, 2011, pp.239-246
  • Takeshi Mori and Shin Ishii, Robust approximation in decomposed reinforcement learning, International Conference on Neural Information Processing (ICONIP), Lecture Notes in Computer Science, 5863, 2009, pp.590-597
  • Takeshi Mori and Shin Ishii, An additive reinforcement learning, International Conference on Artificial Neural Networks (ICANN), Lecture Notes in Computer Science, 5768, 2009, pp.608-617
  • Yu Hiei, Takeshi Mori and Shin Ishii, Self-organized reinforcement learning based on policy gradient in nonstationary environment, International Conference on Artificial Neural Networks (ICANN), Lecture Notes in Computer Science, 5163, 2008, pp.367-376
  • Yuki Taniguchi, Takeshi Mori and Shin Ishii, A continuous internal-state controller for partially observable Markov decision processes, International Conference on Artificial Neural Networks (ICANN), Lecture Notes in Computer Science, 5163, 2008, pp.397-406
  • Tsuyoshi Ueno, Motoaki Kawanabe, Takeshi Mori, Shin-ichi Maeda and Shin Ishii, A semiparametric statistical approach to model-free policy evaluation, International Conference on Machine Learning (ICML), 2008, pp.1072-1079
  • Nobuhito Nanjo, Takeshi Mori and Shin Ishii, An effective reinforcement learning with automatic construction of basis function and sequential approximation, International Symposium on Artificial Life and Robotics (AROB), GS2-3, 2008, pp.662-665
  • Yuki Taniguchi, Takeshi Mori and Shin Ishii, Continuous Internal-State Controller for a Partially Observable Linear Dynamical System, International Symposium on Artificial Life and Robotics (AROB), GS1-1, 2008, pp.674-677
  • Kazuhiro Takeda, Takeshi Mori and Shin Ishii, Active Sampling based on Gaussian Process for Reinforcement Learning, International Symposium on Artificial Life and Robotics (AROB), GS1-2, 2008, pp.678-681
  • Yuki Taniguchi, Takeshi Mori and Shin Ishii, Reinforcement Learning for Cooperative Actions in a Partially Observable Multi-Agent System, International Conference on Artificial Neural Networks (ICANN), Lecture Notes in Computer Science, 4668, 2007, pp.229-238
  • Yutaka Nakamura, Takeshi Mori and Shin Ishii, Natural policy gradient reinforcement learning method for a looper-like robot, International Symposium on Artificial Life and Robotics (AROB), GS3-3, 2006, pp.51-54
  • Yutaka Nakamura, Takeshi Mori and Shin Ishii, An off-policy natural gradient method for a partial observable Markov decision process, International Conference on Artificial Neural Networks (ICANN), Lecture Notes in Computer Science, 3697, 2005, pp.431-436
  • Yutaka Nakamura, Takeshi Mori and Shin Ishii, Natural policy gradient reinforcement learning for a CPG control of a biped robot, International Conference on Parallel Problem Solving from Nature (PPSN), 2004, pp.972-981
  • Takeshi Mori, Yutaka Nakamura, Masa-aki Sato and Shin Ishii, Reinforcement learning for a CPG-driven biped robot, Nineteenth National Conference on Artificial Intelligence (AAAI), 2004, pp.623-630
Workshop Presentations
  • Takeshi Mori, Yutaka Nakamura and Shin Ishii, Efficient Sample Reuse by Off-policy Natural Actor-critic Learning, Advances in Neural Information Processing Systems, Workshop (NIPS Workshop), 2005
  • Takeshi Mori, Yutaka Nakamura and Shin Ishii, Policy-gradient-based actor-critic method for a CPG controller, NAIST/UMN Joint Workshop, 2004
Technical Reports
  • Takeshi Mori, Yutaka Nakamura and Shin Ishii, Off-Policy Natural Actor-Critic, NAIST Technical Report, 2005007, 2005
PhD Thesis
  • Takeshi Mori, Policy-Based Reinforcement Learning, Nara Institute of Science and Technology, 2007
Document Actions