Toward creative problem solving agents: Action discovery through behavior babbling
Creative problem solving (CPS) is the process by which an agent discovers unknown
information about itself and its environment, allowing it to accomplish a previously …
information about itself and its environment, allowing it to accomplish a previously …
EO-MTRNN: evolutionary optimization of hyperparameters for a neuro-inspired computational model of spatiotemporal learning
E Wieser, G Cheng - Biological cybernetics, 2020 - Springer
For spatiotemporal learning with neural networks, hyperparameters are often set manually
by a human expert. This is especially the case with multiple timescale networks that require …
by a human expert. This is especially the case with multiple timescale networks that require …
Robot task learning with motor babbling using pseudo rehearsal
K Kase, A Tateishi, T Ogata - IEEE Robotics and Automation …, 2022 - ieeexplore.ieee.org
The paradigm of deep robot learning from demonstrations allows robots to solve complex
manipulation tasks by capturing motor skills from given demonstrations; however, collecting …
manipulation tasks by capturing motor skills from given demonstrations; however, collecting …
FAD learning: Separate learning for three accelerations-learning for dynamics of boat through motor babbling
A Numakura, S Kato, K Sato… - … on Robotics and …, 2016 - ieeexplore.ieee.org
This paper addresses the modeling and measurement of a small boat. In some fishing tasks,
anchorage is not applicable in order to capture shellfishes or fishes efficiently. Currently …
anchorage is not applicable in order to capture shellfishes or fishes efficiently. Currently …
Vehicle dynamics modeling using FAD learning
Highly precise vehicle dynamics modeling is indispensable for self-driving technology. We
propose a model learning framework, which utilizes FAD (The abbreviation of the capital …
propose a model learning framework, which utilizes FAD (The abbreviation of the capital …
Learning precisely timed feedforward control of the sensor-denied inverted pendulum
TL Mohren, TL Daniel… - IEEE Control Systems …, 2020 - ieeexplore.ieee.org
Time delays due to signal latency, computational complexity, and sensor-denied
environments, pose a critical challenge in both engineered and biological control systems …
environments, pose a critical challenge in both engineered and biological control systems …
Fully automated learning for position and contact force of manipulated object with wired flexible finger joints
K Watanabe, S Nishide, M Gouko, CH Kim - International Conference on …, 2016 - Springer
We discuss about the modeling technology in the object manipulation of the robot arm that is
equipped with flexible finger joints. In recent years, flexible robot fingers are getting attention …
equipped with flexible finger joints. In recent years, flexible robot fingers are getting attention …
Effective input order of dynamics learning tree
In this paper, we discuss about the learning performance of dynamics learning tree (DLT)
while mainly focusing on the implementation on robot arms. We propose an input-order …
while mainly focusing on the implementation on robot arms. We propose an input-order …
力学系学習木による効率的な学習のための階層性を利用した入力ベクトル決定法
濱翔**， **井諒， 高橋城志， 山田浩貴… - 第 78 回全国大会講演 …, 2016 - ipsj.ixsq.nii.ac.jp
論文抄録 力学系学習木により効率的な動作学習法を確立することを目的として,
力学系学習木の持つ階層性を活用した入力ベクトル決定法を提案する. 実験では …
力学系学習木の持つ階層性を活用した入力ベクトル決定法を提案する. 実験では …
ロボティクスと深層学習 (< 特集> ニューラルネットワーク研究のフロンティア)
尾形哲也 - 人工知能, 2016 - jstage.jst.go.jp
また特にロボットと DL との関連で注目を集めているのが, Deep Q-Learning の手法である [Mnih
13]. 本来, 強化学習は, 状態 s と有限数の行動パターン a の組合せからなる Q 値を …
13]. 本来, 強化学習は, 状態 s と有限数の行動パターン a の組合せからなる Q 値を …