LEARNING TO ACT FROM OBSERVATION AND PRACTICE-Reference-Cited by-同舟云学术

LEARNING TO ACT FROM OBSERVATION AND PRACTICE

Published:2004-12 Issue:04 Volume:01 Page:585-611
ISSN:0219-8436
Container-title:International Journal of Humanoid Robotics
language:en
Short-container-title:Int. J. Human. Robot.

Author:

BENTIVEGNA DARRIN C.¹²,ATKESON CHRISTOPHER G.¹³,UDE ALEŠ¹⁴,CHENG GORDON¹²

Affiliation:

1. ATR Computational Neuroscience Laboratories, Department of Humanoid Robotics and Computational Neuroscience, Kyoto, 619-0288, Japan

2. Computational Brain Project, ICORP, Japan Science and Technology Agency, Kyoto, Japan

3. Carnegie Mellon University, Robotics Institute, Pittsburgh, PA, USA

4. Jožef Stefan Institute, Department of Automatics, Biocybernetics and Robotics, Ljubljana, Slovenia

Abstract

We present a method for humanoid robots to quickly learn new dynamic tasks from observing others and from practice. Ways in which the robot can adapt to initial and also changing conditions are described. Agents are given domain knowledge in the form of task primitives. A key element of our approach is to break learning problems up into as many simple learning problems as possible. We present a case study of a humanoid robot learning to play air hockey.

Publisher

World Scientific Pub Co Pte Lt

Subject

Artificial Intelligence,Mechanical Engineering

Link

https://www.worldscientific.com/doi/pdf/10.1142/S0219843604000307

Reference7 articles.

1. Using humanoid robots to study human behavior

2. Neural network-based state prediction for strategy planning of an air hockey robot

3. Control of Planar Rigid Body Sliding with Impacts and Friction

Cited by 25 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Interplay Between Computational Neuroscience and Humanoid Robotics: Past and Future;International Journal of Humanoid Robotics;2023-11-30

2. Predictive reward-prediction errors of climbing fiber inputs integrate modular reinforcement learning with supervised learning;2023-03-13

3. The Robot as Scientist: Using Mental Simulation to Test Causal Hypotheses Extracted from Human Activities in Virtual Reality;2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS);2020-10-24

4. A data-efficient deep learning approach for deployable multimodal social robots;Neurocomputing;2020-07

5. Purposive learning: Robot reasoning about the meanings of human activities;Science Robotics;2019-01-30