Skip to main navigation Skip to search Skip to main content

Applying reinforcement learning to basic routing problem

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Routing jobs to parallel servers is a common and important task in today’s computer and communication systems. As each routing decision affects the jobs arriving later, determining the (near) optimal decisions is non-trivial. In this paper, we apply reinforcement learning techniques to the job routing problem with heterogeneous servers and a general cost structure. We study the convergence of the reinforcement learning to a near-optimal policy (that we can determine by other means), and compare its performance against heuristic policies such as Join-the-Shortest-Queue (JSQ) and Shortest-Expected-Delay (SED).

Original languageEnglish
Title of host publicationQueueing Theory and Network Applications - 13th International Conference, QTNA 2018, Proceedings
EditorsWuyi Yue, Sabine Wittevrongel, Yutaka Takahashi, Tuan Phung-Duc
PublisherSpringer Verlag
Pages238-249
Number of pages12
ISBN (Print)9783319937359
DOIs
Publication statusPublished - 2018
Event13th International Conference on Queueing Theory and Network Applications, QTNA 2018 - Tsukuba, Japan
Duration: 25 Jul 201827 Jul 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10932 LNCS

Conference

Conference13th International Conference on Queueing Theory and Network Applications, QTNA 2018
Country/TerritoryJapan
CityTsukuba
Period25/07/1827/07/18

Bibliographical note

Funding Information: Acknowledgements. This work was supported by the Academy of Finland in the FQ4BD project (grant no. 296206) and by the University of Iceland Research Fund in the RL-STAR project. Publisher Copyright: © Springer International Publishing AG, part of Springer Nature 2018.

Other keywords

  • Job dispatching
  • Machine learning
  • Parallel servers
  • Reinforcement learning
  • Task assignment
  • Value function

Fingerprint

Dive into the research topics of 'Applying reinforcement learning to basic routing problem'. Together they form a unique fingerprint.

Cite this