Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Learning arm's posture control using reinforcement learning and feedback-error-learning.

H Kambara1, J Kim, M Sato

  • 1Dept. of Comput. Intelligence & Syst. Sci., Interdisciplinary Graduate School of Science and Engineering, Tokyo Inst. of Technol., Japan.

Conference Proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference
|February 3, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Safety profile of vonoprazan compared with proton pump inhibitors: insight from a pharmacovigilance study.

Die Pharmazie·2020
Same author

Endpoint stiffness magnitude increases linearly with a stronger power grasp.

Scientific reports·2020
Same author

Increase in Grasp Force Reflects a Desire to Improve Movement Precision.

eNeuro·2019
Same author

Changes in Plasma Lipid Levels in Young Healthy Volunteers by Adding an Extruder-cooked Soy Protein to Conventional Meals.

Bioscience, biotechnology, and biochemistry·2016
Same author

Individualistic weight perception from motion on a slope.

Scientific reports·2016
Same author

Re-biopsy in a patient with IgA nephropathy showing success of treatment with cyclosporin A and angiotensin-II receptor blocker.

Clinical and experimental nephrology·2014
Same journal

Highly Articulated Robotic Probe for Minimally Invasive Surgery.

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference·2010
Same journal

Combined peripheral and cortical neural stimulation in motor neurorehabilitation.

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference·2008
Same journal

Embedded assessment algorithms within home-based cognitive computer game exercises for elders.

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference·2008
Same journal

Digital signal processing of the wideband external pulse recorded during cuff deflation: a new way to measure blood pressure.

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference·2008
Same journal

Immune cells detection of the in vivo rejecting heart in USPIO-enhanced magnetic resonance imaging.

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference·2008
Same journal

wHospital: a web-based application with digital signature for drugs dispensing management.

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference·2007
See all related articles

This study introduces a novel learning model combining the Actor-Critic method and feedback-error-learning for motor control. The model successfully acquires high-performance feedback and feed-forward controllers using only reward signals.

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Robotics

Background:

  • The Actor-Critic method is a key reinforcement learning framework relevant to basal ganglia function.
  • Feedback-error-learning is a model of cerebellar motor learning that uses feedback controller output as error signals.
  • Previous feedback-error-learning models used fixed-gain feedback controllers.

Purpose of the Study:

  • To propose a novel learning model integrating Actor-Critic and feedback-error-learning.
  • To utilize the Actor-Critic method for generating the feedback controller within the feedback-error-learning architecture.
  • To demonstrate the model's capability in acquiring effective motor controllers.

Main Methods:

  • The proposed model combines the Actor-Critic method with the feedback-error-learning scheme.

Related Experiment Videos

  • The Actor-Critic method is employed to derive the feedback controller.
  • The learning architecture aims to acquire a feed-forward controller using the feedback controller's output as an error signal.

    • Main Results:

    • The integrated learning model was applied to an arm's posture control task.
    • The model successfully acquired both feedback and feed-forward controllers.
    • High-performance control was achieved using only scalar reward values.

    Conclusions:

    • The proposed learning model effectively integrates Actor-Critic and feedback-error-learning.
    • This approach enables the acquisition of sophisticated motor controllers.
    • The model demonstrates efficient learning from scalar reward signals, applicable to robotic systems and neuroscience.