Conference paper Open Access

Optimizing QoE and Cost in a 3D Immersive Media Platform: A Reinforcement Learning Approach

Panagiotis Athanasoulis; Emmanouil Christakis; Konstantinos Konstantoudakis; Petros Drakoulis; Stamatia Rizou; Avi Weit; Alexandros Doumanoglou; Nikolaos Zioulis; Dimitrios Zarpalas


MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nam##2200000uu#4500</leader>
  <datafield tag="041" ind1=" " ind2=" ">
    <subfield code="a">eng</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">immersive media</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">cognitive network optimizer</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">reinforcement learning</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">5g</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">5g media</subfield>
  </datafield>
  <controlfield tag="005">20200423202020.0</controlfield>
  <controlfield tag="001">3761854</controlfield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Visual Computing Lab (VCL), Information Technologies Institute (ITI), Centre for Research and Technology - Hellas (CERTH), Thessaloniki, Greece</subfield>
    <subfield code="a">Emmanouil Christakis</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Visual Computing Lab (VCL), Information Technologies Institute (ITI), Centre for Research and Technology - Hellas (CERTH), Thessaloniki, Greece</subfield>
    <subfield code="0">(orcid)0000-0001-5092-8796</subfield>
    <subfield code="a">Konstantinos Konstantoudakis</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Visual Computing Lab (VCL), Information Technologies Institute (ITI), Centre for Research and Technology - Hellas (CERTH), Thessaloniki, Greece</subfield>
    <subfield code="0">(orcid)0000-0003-3434-3290</subfield>
    <subfield code="a">Petros Drakoulis</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Singular Logic S.A., Athens, Greece</subfield>
    <subfield code="a">Stamatia Rizou</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">IBM Research, Haifa, Israel</subfield>
    <subfield code="a">Avi Weit</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Visual Computing Lab (VCL), Information Technologies Institute (ITI), Centre for Research and Technology - Hellas (CERTH), Thessaloniki, Greece</subfield>
    <subfield code="0">(orcid)0000-0002-4337-1720</subfield>
    <subfield code="a">Alexandros Doumanoglou</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Visual Computing Lab (VCL), Information Technologies Institute (ITI), Centre for Research and Technology - Hellas (CERTH), Thessaloniki, Greece</subfield>
    <subfield code="0">(orcid)0000-0002-7898-9344</subfield>
    <subfield code="a">Nikolaos Zioulis</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Visual Computing Lab (VCL), Information Technologies Institute (ITI), Centre for Research and Technology - Hellas (CERTH), Thessaloniki, Greece</subfield>
    <subfield code="a">Dimitrios Zarpalas</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">781740</subfield>
    <subfield code="z">md5:33f3abda084213aa9f77d028833c215c</subfield>
    <subfield code="u">https://zenodo.org/record/3761854/files/Optimizing QoE and Cost in a 3D Immersive Media Platform - A Reinforcement Learning Approach - camera ready.pdf</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2020-04-22</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">openaire</subfield>
    <subfield code="o">oai:zenodo.org:3761854</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="u">Singular Logic S.A., Athens, Greece</subfield>
    <subfield code="a">Panagiotis Athanasoulis</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Optimizing QoE and Cost in a 3D Immersive Media Platform: A Reinforcement Learning Approach</subfield>
  </datafield>
  <datafield tag="536" ind1=" " ind2=" ">
    <subfield code="c">761699</subfield>
    <subfield code="a">Programmable edge-to-cloud virtualization fabric for the 5G Media industry</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u">https://creativecommons.org/licenses/by/4.0/legalcode</subfield>
    <subfield code="a">Creative Commons Attribution 4.0 International</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;Recent advances in media-related technologies, including&amp;nbsp;capturing and processing, have facilitated novel forms&amp;nbsp;of 3D media content, increasing the degree of user immersion.&amp;nbsp;In order to ensure these technologies can readily support the&amp;nbsp;rising demand for more captivating entertainment, both the&amp;nbsp;production and delivery mechanisms should be transformed to&amp;nbsp;support the application of media or network-related optimizations&amp;nbsp;and refinements on-the-fly. Network peculiarities deriving&amp;nbsp;from geographic and other factors make it difficult for a greedy&amp;nbsp;or a supervised machine learning algorithm to successfully foresee&amp;nbsp;the need for reconfiguration of the content production or delivery&amp;nbsp;procedures. For these reasons, Reinforcement Learning (RL)&amp;nbsp;approaches have lately gained popularity as partial information&amp;nbsp;on the environment is enough for an algorithm to begin its&amp;nbsp;training and converge to an optimal policy. The contribution&amp;nbsp;of this work is a Cognitive Network Optimizer (CNO) in the&amp;nbsp;form of an RL agent, designed to perform corrective actions on&amp;nbsp;both the production and consumption ends of an immersive 3D&amp;nbsp;media platform, depending on a collection of real-time monitoring&amp;nbsp;parameters, including infrastructure, application-level and quality&amp;nbsp;of experience (QoE) metrics. Our work demonstrates CNO&amp;nbsp;approaches with different foci, i.e., a greedy maximization of&amp;nbsp;the users&amp;rsquo; QoE, a QoE-focused RL approach and a combined&amp;nbsp;QoE-and-Cost RL approach.&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="773" ind1=" " ind2=" ">
    <subfield code="n">doi</subfield>
    <subfield code="i">isVersionOf</subfield>
    <subfield code="a">10.5281/zenodo.3761853</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.5281/zenodo.3761854</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">publication</subfield>
    <subfield code="b">conferencepaper</subfield>
  </datafield>
</record>
74
50
views
downloads
All versions This version
Views 7474
Downloads 5050
Data volume 39.1 MB39.1 MB
Unique views 6363
Unique downloads 4444

Share

Cite as