Publications

• Sorted by Date • Classified by Publication Type • Classified by Research Category •

Distributed Influence-Augmented Local Simulators for Parallel MARL in Large Networked Systems

Miguel Suau, Jinke He, Mustafa Mert Çelikok, Matthijs T. J. Spaan, and Frans A. Oliehoek. Distributed Influence-Augmented Local Simulators for Parallel MARL in Large Networked Systems. In Advances in Neural Information Processing Systems 35, December 2022.

Download

pdf [658.0kB]  

Abstract

Due to its high sample complexity, simulation is, as of today, critical for the successful application of reinforcement learning. Many real-world problems, however, exhibit overly complex dynamics, making their full-scale simulation computationally slow. In this paper, we show how to factorize large networked systems of many agents into multiple local regions such that we can build separate simulators that run independently and in parallel. To monitor the influence that the different local regions exert on one another, each of these simulators is equipped with a learned model that is periodically trained on real trajectories. Our empirical results reveal that distributing the simulation among different processes not only makes it possible to train large multi-agent systems in just a few hours but also helps mitigate the negative effects of simultaneous learning.

BibTeX Entry

@inproceedings{Suau22NeurIPS,
    title=      {Distributed Influence-Augmented Local Simulators for 
                Parallel {MARL} in Large Networked Systems},
    author=     {Miguel Suau and Jinke He and  {\c C}elikok, Mustafa Mert
                and Matthijs T. J. Spaan and Frans A. Oliehoek},
    booktitle=  NIPS35,
    year=       2022,
    month=      dec,
    url={https://openreview.net/forum?id=lKFOwaYNQlb},
    abstract={
        Due to its high sample complexity, simulation is, as of today, critical
        for the successful application of reinforcement learning. Many
        real-world problems, however, exhibit overly complex dynamics, making
        their full-scale simulation computationally slow. In this paper, we
        show how to factorize large networked systems of many agents into
        multiple local regions such that we can build separate simulators that
        run independently and in parallel. To monitor the influence that the
        different local regions exert on one another, each of these simulators
        is equipped with a learned model that is periodically trained on real
        trajectories. Our empirical results reveal that distributing the
        simulation among different processes not only makes it possible to
        train large multi-agent systems in just a few hours but also helps
        mitigate the negative effects of simultaneous learning.
    }
}

Generated by bib2html.pl (written by Patrick Riley) on Tue Nov 05, 2024 16:13:37 UTC