Interactive Learning of Temporal Features for Control: Shaping Policies and State Representations From Human Feedback

Rodrigo Perez-Dattari; Carlos Celemin; Giovanni Franzese; Javier Ruiz-del-Solar; Jens Kober

doi:10.1109/MRA.2020.2983649

Interactive Learning of Temporal Features for Control: Shaping Policies and State Representations From Human Feedback

Rodrigo Perez-Dattari, Carlos Celemin, Giovanni Franzese, Javier Ruiz-del-Solar, Jens Kober

Learning & Autonomous Control

Research output: Contribution to journal › Article › Scientific › peer-review

8 Citations (Scopus)

126 Downloads (Pure)

Abstract

Current ongoing industry revolution demands more flexible products, including robots in household environments and medium-scale factories. Such robots should be able to adapt to new conditions and environments and be programmed with ease. As an example, let us suppose that there are robot manipulators working on an industrial production line and that they need to perform a new task. If these robots were hard coded, it could take days to adapt them to the new settings, which would stop production at the factory. Robots that non-expert humans could easily program would speed up the process considerably.

Original language	English
Pages (from-to)	46-54
Journal	IEEE Robotics and Automation Magazine
Volume	27
Issue number	2
DOIs	https://doi.org/10.1109/MRA.2020.2983649
Publication status	Published - 2020

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care

Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Artificial neural networks
Service robots
Task analysis
Feature extraction
Training
Computer architecture

Access to Document

10.1109/MRA.2020.2983649

Interactive_Learning_of_Temporal_Features_for_Control_Shaping_Policies_and_State_Representations_From_Human_FeedbackFinal published version, 2.36 MB

Cite this

@article{26fc3d8662e54dc98cac5cd2c7e8536a,

title = "Interactive Learning of Temporal Features for Control: Shaping Policies and State Representations From Human Feedback",

abstract = "Current ongoing industry revolution demands more flexible products, including robots in household environments and medium-scale factories. Such robots should be able to adapt to new conditions and environments and be programmed with ease. As an example, let us suppose that there are robot manipulators working on an industrial production line and that they need to perform a new task. If these robots were hard coded, it could take days to adapt them to the new settings, which would stop production at the factory. Robots that non-expert humans could easily program would speed up the process considerably.",

keywords = "Artificial neural networks, Service robots, Task analysis, Feature extraction, Training, Computer architecture",

author = "Rodrigo Perez-Dattari and Carlos Celemin and Giovanni Franzese and Javier Ruiz-del-Solar and Jens Kober",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.",

year = "2020",

doi = "10.1109/MRA.2020.2983649",

language = "English",

volume = "27",

pages = "46--54",

journal = "IEEE Robotics and Automation Magazine",

issn = "1070-9932",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

number = "2",

}

TY - JOUR

T1 - Interactive Learning of Temporal Features for Control

T2 - Shaping Policies and State Representations From Human Feedback

AU - Perez-Dattari, Rodrigo

AU - Celemin, Carlos

AU - Franzese, Giovanni

AU - Ruiz-del-Solar, Javier

AU - Kober, Jens

N1 - Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2020

Y1 - 2020

N2 - Current ongoing industry revolution demands more flexible products, including robots in household environments and medium-scale factories. Such robots should be able to adapt to new conditions and environments and be programmed with ease. As an example, let us suppose that there are robot manipulators working on an industrial production line and that they need to perform a new task. If these robots were hard coded, it could take days to adapt them to the new settings, which would stop production at the factory. Robots that non-expert humans could easily program would speed up the process considerably.

AB - Current ongoing industry revolution demands more flexible products, including robots in household environments and medium-scale factories. Such robots should be able to adapt to new conditions and environments and be programmed with ease. As an example, let us suppose that there are robot manipulators working on an industrial production line and that they need to perform a new task. If these robots were hard coded, it could take days to adapt them to the new settings, which would stop production at the factory. Robots that non-expert humans could easily program would speed up the process considerably.

KW - Artificial neural networks

KW - Service robots

KW - Task analysis

KW - Feature extraction

KW - Training

KW - Computer architecture

UR - http://www.scopus.com/inward/record.url?scp=85083780719&partnerID=8YFLogxK

U2 - 10.1109/MRA.2020.2983649

DO - 10.1109/MRA.2020.2983649

M3 - Article

AN - SCOPUS:85083780719

SN - 1070-9932

VL - 27

SP - 46

EP - 54

JO - IEEE Robotics and Automation Magazine

JF - IEEE Robotics and Automation Magazine

IS - 2

ER -

Interactive Learning of Temporal Features for Control: Shaping Policies and State Representations From Human Feedback

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this