Trabajo presentado a la Twenty-Ninth AAAI Conference on Artificial Intelligence (AAAI-15) celebrada en Austin, Texas (US) del 25 al 30 de enero de 2015. ; Reinforcement learning (RL) is a common paradigm for learning tasks in robotics. However, a lot of exploration is usually required, making RL too slow for high-level tasks. We present V-MIN, an algorithm that integrates teacher demonstrations with RL to learn complex tasks faster. The algorithm combines active demonstration requests and autonomous exploration to find policies yielding rewards higher than a given threshold Vmin. This threshold sets the degree of quality with which the robot is expected to complete the task, thus allowing the user to either opt for very good policies that require many learning experiences, or to be more permissive with sub-optimal policies that are easier to learn. The threshold can also be increased online to force the system to improve its policies until the desired behavior is obtained. Furthermore, the algorithm generalizes previously learned knowledge, adapting well to changes. The performance of V-MIN has been validated through experimentation, including domains from the international planning competition. Our approach achieves the desired behavior where previous algorithms failed. ; This work is supported by CSIC project MANIPlus 201350E102 and by the Spanish Ministry of Science and Innovation under project PAU+ DPI2011-27510. D. Martínez is also supported by the Spanish Ministry of Education, Culture and Sport via a FPU doctoral grant (FPU12-04173). ; Peer Reviewed


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    V-MIN: Efficient reinforcement learning through demonstrations and relaxed reward demands



    Erscheinungsdatum :

    2015-01-01



    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch


    Klassifikation :

    DDC:    629



    V-MIN: efficient reinforcement learning through demonstrations and relaxed reward demands

    Martínez Martínez, David / Alenyà Ribas, Guillem / Torras, Carme | BASE | 2015

    Freier Zugriff

    Relational reinforcement learning with guided demonstrations

    Martínez Martínez, David / Alenyà Ribas, Guillem / Torras, Carme | BASE | 2017

    Freier Zugriff


    Human Demonstrations for Fast and Safe Exploration in Reinforcement Learning

    Schonebaum, Gerben / Junell, Jaime / Van Kampen, Erik-Jan | AIAA | 2017


    Low Speed Airship Control using Reinforcement Learning and Expert Demonstrations

    Daskiran, Onur / Huff, Brian / Dogan, Atilla | AIAA | 2017