This paper presents a training method for increasing performance of reinforcement learning agents. The method is named Environment Adversarial Reinforcement Learning. The method requires the reinforcement learning environment to be parameterizeable. Over the course of training, environment parameters are updated in a direction of increasing difficulty for the agent. The direction for these updates is found using a performance prediction network trained on data from tests of the agent under varying environment parameters. The method was tested on a CartPole environment. A 28-58\% improvement in mean return was found when comparing performance to a baseline reinforcement learning algorithm on both easy and hard versions of the task.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen


    Exportieren, teilen und zitieren



    Titel :

    Environment Adversarial Reinforcement Learning


    Beteiligte:
    John R Cooper (Autor:in)

    Kongress:

    AIAA SciTech Forum and Exposition ; 2024 ; Orlando, FL, US


    Medientyp :

    Sonstige


    Format :

    Keine Angabe


    Sprache :

    Englisch





    CERTIFIED ADVERSARIAL ROBUSTNESS FOR DEEP REINFORCEMENT LEARNING

    LUETJENS BJOERN MALTE / EVERETT MICHAEL F / HOW JONATHAN P et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Learning Aircraft Pilot Skills by Adversarial Inverse Reinforcement Learning

    Suzuki, Kaito / Uemura, Tsuneharu / Tsuchiya, Takeshi et al. | Springer Verlag | 2024


    Adversarial Proximal Policy Optimisation for Robust Reinforcement Learning

    Ince, Bilkan / Shin, Hyo-Sang / Tsourdos, Antonios | AIAA | 2024


    Modeling Driver Behavior using Adversarial Inverse Reinforcement Learning

    Sackmann, Moritz / Bey, Henrik / Hofmann, Ulrich et al. | IEEE | 2022