Publications

Sorted by DateClassified by Publication TypeClassified by Research Category

Influence-Augmented Local Simulators: A Scalable Solution for Fast Deep RL in Large Networked Systems

Miguel Suau, Jinke He, Matthijs T. J. Spaan, and Frans A. Oliehoek. Influence-Augmented Local Simulators: A Scalable Solution for Fast Deep RL in Large Networked Systems. In Proceedings of the 39th International Conference on Machine Learning (ICML), pp. 20604–20624, 2022.

Download

pdf [1.8MB]  

Abstract

Learning effective policies for real-world problems is still an open challenge for the field ofreinforcement learning (RL). The main limitation being the amount of data needed and the pace at which that data can be obtained. In this paper, we study how to build lightweight simulators of complicated systems that can run sufficiently fast for deep RL to be applicable. We focus on domains where agents interact with a reduced portion of a larger environment while still being affected by the global dynamics. Our method combines the use of local simulators with learned models that mimic the influence of the global system. The experiments reveal that incorporating this idea into the deep RL workflow can considerably accelerate the training process and presents several opportunities for the future.

BibTeX Entry

@inproceedings{Suau22ICML,
    title =     {Influence-Augmented Local Simulators: A Scalable Solution for Fast Deep {RL} in Large Networked Systems},
    author =    {Miguel Suau and Jinke He and Matthijs T. J. Spaan and Frans A. Oliehoek},
    booktitle = ICML22,
    year =      2022,
    pages =     {20604--20624},
    OPTseries = {Proceedings of Machine Learning Research},
    keywords =   {refereed},
    abstract = {
Learning effective policies for real-world problems is still an open challenge for the field of
reinforcement learning (RL). The main limitation being the amount of data needed and the pace at 
which that data can be obtained. In this paper, we study how to build lightweight simulators of 
complicated systems that can run sufficiently fast for deep RL to be applicable. We focus on 
domains where agents interact with a reduced portion of a larger environment while still being 
affected by the global dynamics. Our method combines the use of local simulators with learned 
models that mimic the influence of the global system. The experiments reveal that incorporating 
this idea into the deep RL workflow can considerably accelerate the training process and 
presents several opportunities for the future.
    }
}

Generated by bib2html.pl (written by Patrick Riley) on Sat Nov 12, 2022 12:13:03 UTC