BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211207T055403Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20211116T083000
DTEND;TZID=America/Chicago:20211116T170000
UID:submissions.supercomputing.org_SC21_sess278_rpost145@linklings.com
SUMMARY:Optimizing and Extending the Functionality of EXARL for Scalable R
 einforcement Learning
DESCRIPTION:Posters, Research Posters\n\nOptimizing and Extending the Func
 tionality of EXARL for Scalable Reinforcement Learning\n\nChenna, Cosburn,
  Ezeobi, Moraru, Lim...\n\nEasily eXtendable Architecture for Reinforcemen
 t Learning (EXARL) is a scalable reinforcement learning framework, part of
  the Exascale Computing Project (ECP) funded ExaLearn program, designed to
  facilitate reinforcement learning (RL) research for complex scientific en
 vironments. In RL, agents are algorithms that interact and learn from an e
 nvironment, such as a game or a scientific simulation, with an aim to maxi
 mize reward. We improve the existing EXARL framework by optimizing and ext
 ending the current functionality by incorporating additional agents, such 
 as Advantage Actor-Critic (A2C/A3C) and Twin Delayed Deep Deterministic Po
 licy Gradient (TD3) to the framework, as well as improve upon these and ex
 isting agents using v-trace and prioritized experience replay. We also imp
 rove the scalability of the framework by optimizing communication across m
 ultiple learners and between actors and learners, as well as accelerate th
 e data generation pipeline of the Deep Q-Network (DQN) agent by leveraging
  environment processes.\n\nRegistration Category: Tech Program Reg Pass, E
 xhibit Hall Only
END:VEVENT
END:VCALENDAR