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:20251016T135148Z
LOCATION:Riverside East
DTSTART;TZID=America/Chicago:20251014T173000
DTEND;TZID=America/Chicago:20251014T183000
UID:HFESAM_ASPIRE 2025_sess131_POST508@linklings.com
SUMMARY:When Forgetting is Learning: Human-Inspired Memory Management in a
  Policy Capturing AI System
DESCRIPTION:Léandre Lavoie-Hudon and Coralie Bureau (Université Laval), Da
 niel Lafond (Thales), and Sébastien Tremblay (Université Laval)\n\nArtific
 ial intelligence (AI) systems must adapt quickly to dynamic and interactiv
 e environments, but maintaining relevance over time remains a challenge. I
 nspired by the adaptive advantages of human forgetting, this study investi
 gates the integration of a forgetting function into an AI system. We imple
 mented this mechanism as a training window within the Cognitive Shadow (CS
 ) system, an AI designed to learn and emulate human decision models for cl
 assification tasks. This training window hyperparameter – applicable to an
 y supervised machine learning algorithm – helps address the issue of conce
 pt drift by prioritizing recent information. The effectiveness of this add
 ition was tested with a simple strategy game similar in dynamics to rock-p
 aper-scissors. Each participant played three sessions of 12 battles, with 
 each consisting of five rounds, against an AI opponent. CS was trained dur
 ing Session 1 and became active in Sessions 2 and 3. Analyses showed that 
 including the training window significantly improved predictive accuracy o
 f CS in both Sessions 2 and 3 by emphasizing recent, relevant data and fil
 tering out noise. These findings highlight the potential of incorporating 
 human-inspired forgetting mechanisms to enhance AI performance in interact
 ive and dynamic environments, with implications for real-time decision-mak
 ing support systems and interactive applications.\n\n
END:VEVENT
END:VCALENDAR