BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160550Z
LOCATION:Poster Module
DTSTART;TZID=America/New_York:20201119T083000
DTEND;TZID=America/New_York:20201119T170000
UID:submissions.supercomputing.org_SC20_sess341_spostg106@linklings.com
SUMMARY:Multi-Agent Meta Reinforcement Learning for Packet Routing in Dyna
 mic Network Environments
DESCRIPTION:ACM Student Research Competition: Graduate Poster, ACM Student
  Research Competition: Undergraduate Poster, Posters\n\nMulti-Agent Meta R
 einforcement Learning for Packet Routing in Dynamic Network Environments\n
 \nSun\n\nTraffic optimization challenges, such as flow scheduling and comp
 letion time reducing, are difficult online decision-making problems in wid
 e area networks. Previous works apply heuristics that rely on full knowled
 ge of the system to design optimization algorithms. In this work, we explo
 re building a model-free approach, applying multi-agent meta reinforcement
  learning to solve complex online control problem that generates optimal p
 aths to reroute traffic. Focusing on decentralized solutions, our experime
 nt aims to efficiently minimize the average packet completion time while r
 educing packet loss across complex network topologies. To evaluate, we tes
 t with a static topology and dynamically changing network topologies and c
 ompare results to the classical shorted path algorithm.\n\nTag: Student Pr
 ogram\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg Pass
END:VEVENT
END:VCALENDAR