Learning Centre

Multi-agent deep learning for simultaneous optimization for time and energy in distributed routing system

 |  Login

Show simple item record

dc.contributor Aalto-yliopisto fi
dc.contributor Aalto University en
dc.contributor.author Mukhutdinov, Dmitry
dc.contributor.author Filchenkov, Andrey
dc.contributor.author Shalyto, Anatoly
dc.contributor.author Vyatkin, Valeriy
dc.date.accessioned 2019-02-25T08:49:13Z
dc.date.available 2019-02-25T08:49:13Z
dc.date.issued 2019-05-01
dc.identifier.citation Mukhutdinov , D , Filchenkov , A , Shalyto , A & Vyatkin , V 2019 , ' Multi-agent deep learning for simultaneous optimization for time and energy in distributed routing system ' , Future Generation Computer Systems , vol. 94 , pp. 587-600 . https://doi.org/10.1016/j.future.2018.12.037 en
dc.identifier.issn 0167-739X
dc.identifier.other PURE UUID: 8b4fe889-78a5-4de6-94c4-007587b8504a
dc.identifier.other PURE ITEMURL: https://research.aalto.fi/en/publications/8b4fe889-78a5-4de6-94c4-007587b8504a
dc.identifier.other PURE LINK: http://www.scopus.com/inward/record.url?scp=85059178802&partnerID=8YFLogxK
dc.identifier.other PURE FILEURL: https://research.aalto.fi/files/31644647/ELEC_Mukhutdinov_Multi_Agent_Deep_Learning_FutGenComSup_94_587_2019.pdf
dc.identifier.uri https://aaltodoc.aalto.fi/handle/123456789/36810
dc.description.abstract Packet routing problem most commonly emerges in the context of computer networks, thus the majority of routing algorithms existing nowadays is designed specifically for routing in computer networks. However, in the logistics domain, many problems can be formulated in terms of packet routing, e.g. in automated traffic routing or material handling systems. In this paper, we propose an algorithm for packet routing in such heterogeneous environments. Our approach is based on deep reinforcement learning networks combined with link-state protocol and preliminary supervised learning. Similarly to most routing algorithms, the proposed algorithm is a distributed one and is designed to run on a network constructed from interconnected routers. Unlike most other algorithms, proposed one views routers as learning agents, representing the routing problem as a multi-agent reinforcement learning problem. Modeling each router as a deep neural network allows each router to account for heterogeneous data about its environment, allowing for optimization of more complex cost functions, like in case of simultaneous optimization of bag delivery time and energy consumption in a baggage handling system. We tested the algorithm using manually constructed simulation models of computer network and baggage handling system. It outperforms state-of-the-art routing algorithms. en
dc.format.extent 14
dc.format.extent 587-600
dc.format.mimetype application/pdf
dc.language.iso en en
dc.relation.ispartofseries Future Generation Computer Systems en
dc.relation.ispartofseries Volume 94 en
dc.rights openAccess en
dc.title Multi-agent deep learning for simultaneous optimization for time and energy in distributed routing system en
dc.type A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä fi
dc.description.version Peer reviewed en
dc.contributor.department St. Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO)
dc.contributor.department Department of Electrical Engineering and Automation
dc.subject.keyword Deep reinforcement learning
dc.subject.keyword Distributed systems
dc.subject.keyword Multi-agent learning
dc.identifier.urn URN:NBN:fi:aalto-201902251967
dc.identifier.doi 10.1016/j.future.2018.12.037
dc.date.embargo info:eu-repo/date/embargoEnd/2020-12-28

Files in this item

Files Size Format View

There are no open access files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search archive

Advanced Search

article-iconSubmit a publication