中村 泰 NAKAMURA Yutaka

ID:1000070403334

株式会社国際電気通信基礎技術研究所:大阪大学大学院基礎工学研究科 Advanced Telecommunications Research Institute International (ATR):Graduate School of Engineering Science, Osaka University (2015年 CiNii収録論文より)

Search authors sharing the same name

Articles:  1-18 of 18

  • Reconstruction of three-dimensional multi-finger movements from magnetoencephalography with a state transition model  [in Japanese]

    AZUMA Yoshiki , SHIKAUCHI Yumi , Nakamura Yutaka , HIRAYAMA Jun-ichiro , ISHII Shin

    We proposed a new reconstruction model to reconstruct multi-finger movements from MEG signals. We extracted features of brain activities by ICA and classified brain states with HMM. The new reconstruc …

    IEICE technical report. Neurocomputing 114(515), 199-204, 2015-03-16

  • Reconstruction of three-dimensional multi-finger movements from magnetoencephalography with a state transition model  [in Japanese]

    AZUMA Yoshiki , SHIKAUCHI Yumi , Nakamura Yutaka , HIRAYAMA Jun-ichiro , ISHII Shin

    We proposed a new reconstruction model to reconstruct multi-finger movements from MEG signals. We extracted features of brain activities by ICA and classified brain states with HMM. The new reconstruc …

    IEICE technical report. ME and bio cybernetics 114(514), 199-204, 2015-03-16

    Ichushi Web 

  • Policy gradient method for a policy function with probabilistic parameters  [in Japanese]

    NAKAMURA Yutaka

    Stochastic policy gradient methods are a type of reinforcement learning method, where the parameter of the policy parameter is updated according to the gradient with respect to called policy gradient. …

    IEICE technical report 107(542), 343-348, 2008-03-05

    References (18)

  • Acquiring vermicular motion of a Looper-like robot based on the CPG-Actor-Critic method  [in Japanese]

    MAKINO Kenji , NAKAMURA Yutaka , SHIBATA Tomohiro , ISHII Shin

    Adaptability to the environment is crucial for mobile robots, because the circumstance, including the body of the robot, may change. A robot with a large number of degrees of freedom possesses potenti …

    IEICE technical report 106(588), 203-208, 2007-03-07

    References (16)

  • Learning of a robust controller for a biped robot based on a sample-reuse reinforcement learning method  [in Japanese]

    UENO Tsuyoshi , NAKAMURA Yutaka , TAKUMA Takashi , SHIBATA Tomohiro , HOSODA Koh , ISHII Shin

    Recently, many researchers on humanoid robotics are interested in Quasi-Passive Dynamic Walking (Quasi-PDW), which is similar to human walking. It is desirable that control parameters in Quasi-PDW are …

    IEICE technical report 106(588), 197-202, 2007-03-07

    References (12)

  • Feature Extraction for Decision-Theoretic Planning in Partially Observable Stochastic Domains  [in Japanese]

    FUJITA Hajime , NAKAMURA Yutaka , ISHII Shin

    We propose a feature extraction technique for decision-theoretic planning problems in partially observable stochastic domains and show a novel approach for solving them. To maximize an expected future …

    IEICE technical report 106(102), 13-18, 2006-06-09

    References (23)

  • Efficient Sample Reuse by Natural Actor-Critic Learning Based on Importance Sampling  [in Japanese]

    MORI Takeshi , NAKAMURA Yutaka , ISHII Shin

    近年,方策こう配法に基づく強化学習法の一種であるNatural Actor-Critic法(NAC)が提案された.この手法は,actorの学習に自然方策こう配法,criticの学習にLSTD-Q(λ)法を用いたもので,高次元の力学系に対する比較的効率の良いモデルフリー強化学習法として注目されている.しかしながら,NACは方策オン型,すなわち現在の方策に依存した学習法であることにより,二つの問題点が …

    The IEICE transactions on information and systems 89(5), 954-966, 2006-05-01

    References (22) Cited by (2)

  • An application of reinforcement learning with consideration of modeling error  [in Japanese]

    TOKITA Yoichi , NAKAMURA Yutaka , YOSHIMOTO Junichiro , ISHII Shin

    Because reinforcement learning (RL) methods have an advantage such that a control rule can be obtained autonomously without any knowledge of the target system. RL methods have been successfully applie …

    IEICE technical report 105(659), 19-24, 2006-03-17

    References (10)

  • Learning of Qusasi-Passive Dynamic Walking by a Stochastic Policy Gradient Method  [in Japanese]

    HITOMI Kentarou , SHIBATA Tomohiro , NAKAMURA Yutaka , ISHII Shin

    A class of biped locomotion called Passive Dynamic Walking (PDW) has been recognized to be efficient in energy consumption and a key to understand human walking. Although PDW is sensitive to the initi …

    IEICE technical report. Neurocomputing 105(211), 31-36, 2005-07-20

    References (13)

  • Off-Policy Natural Actor-Critic

    MORI Takeshi , NAKAMURA Yutaka , ISHII Shin

    Recently-developed Natural Actor-Critic (NAC), which employs natural policy gradient learning for the actor and LSTD-Q(λ) for the critic, has provided a good model-free reinforcement learning scheme a …

    IEICE technical report. Neurocomputing 105(211), 25-30, 2005-07-20

    References (18)

  • Reinforcement Learning Based on a Policy Gradient Method for a Biped Locomotion  [in Japanese]

    MORI Takeshi , NAKAMURA Yutaka , ISHII Shin

    近年, 「方策こう配法に基づくactor-critic法」が開発された. この手法では, 方策パラメータによって決まる基底関数の線形和によって価値関数の近似を行うため, 価値関数の学習が比較的容易であり, ロボット制御などの大きな状態空間をもつ実問題に対して有用であると考えられる. 我々は以前, 生物を規範とした運動制御機構であるCPGコントローラに対する強化学習モデルとして, CPG-actor …

    The IEICE transactions on information and systems Pt. 2 88(6), 1080-1089, 2005-06

    Cited by (2)

  • Control of Real Acrobot by Learning the Switching Rule of Multiple Controllers  [in Japanese]

    NISHIMURA Masaya , YOSHIMOTO Junichiro , TOKITA Yoichi , NAKAMURA Yutaka , ISHII Shin

    アクロボットは2リンク2関節からなる劣駆動マニピュレータであり, その制御設計は困難な非線形問題であることが知られている.本研究では, 制御理論及び機械学習の分野で得られた知見を統合することにより, 実アクロボットに対する適応的制御設計法を提案する.提案手法では, まず, システム同定法によって近似されたシステム方程式に基づいて部分問題を解くための制御器を複数個設計する.そして, 各制御器を適用す …

    The IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences (Japanese edition) A 88(5), 646-657, 2005-05-01

    References (19) Cited by (3)

  • An off-policy reinforcement learning method based on a natural policy gradient method  [in Japanese]

    NAKAMURA Yutaka , ISHII Shin

    There has been a problem called "exploration-exploitation problem" in the field of reinforcement learning. That is, the agent must determine whether to explore better action which may not necessarily …

    IEICE technical report. Neurocomputing 104(759), 131-136, 2005-03-22

    References (19)

  • A reinforcement learning for a policy involving value-directed internal state  [in Japanese]

    NAKAMURA Yutaka , ISHII Shin

    There are many studies on partially observable Markov decision processes, which employ "belief state" that represents the state of the environment, in order to estimate the value function. However, it …

    IEICE technical report. Neurocomputing 104(140), 29-34, 2004-06-18

    References (13)

  • Reinforcement learning based on a policy gradient method for biped locomotion  [in Japanese]

    MORI Takeshi , NAKAMURA Yutaka , ISHII Shin

    Recently, an actor-critic method utilizing a lower dimensional projection of the value function based on a policy gradient method has been proposed. In this actor-critic method, the approximation of t …

    IEICE technical report. Neurocomputing 103(734), 73-78, 2004-03-12

    References (11)

  • Reinforcement Learning for Rhythmic Movements Using a Neural Oscillator Network  [in Japanese]

    NAKAMURA Yutaka , SATO Masa-aki , ISHII Shin

    歩行などの生物の運動は,周期的な信号を生成する中枢パターン生成器(CPG)と呼ばれる神経回路によって制御されていることが示唆されている.このような生物の制御機構を参考にして,周期的な運動に対するCPGを用いた制御法の研究が行われてきた.我々は,CPGコントローラを用いた自律的な学習制御の枠組みとして,CPG-actor-criticモデルと呼ばれる新しい強化学習法を提案する.我々は2足歩行ロボット …

    The Transactions of the Institute of Electronics,Information and Communication Engineers. 87(3), 893-902, 2004-03-01

    References (14) Cited by (3)

  • An analysis of human pointing motion by online Bayes method  [in Japanese]

    NAKAMURA Yutaka , OBA Shigeyuki , YOSHIMOTO Junichiro , ISHII Shin

    Human motions have been studied for gesture recognition, computer graphics, robot control with referring human motions, and so on. A linear state space model is often used for modeling and analyzing s …

    IEICE technical report. Neurocomputing 102(729), 149-153, 2003-03-10

    References (10)

  • Acquisition of locomotion by reinforcement learning using neural oscillator network  [in Japanese]

    NAKAMURA Yutaka , ISHII Shin , SATO Masa-aki

    Animal rhythmic movements such as locomotion are considered to be controlled by neural circuits called central pattern generators (CPGs), which generate oscillatory signals. Motivated by such a contro …

    IEIC Technical Report (Institute of Electronics, Information and Communication Engineers) 101(735), 183-190, 2002-03-11

    References (7) Cited by (1)

Page Top