BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250822T115806Z
LOCATION:Room 5.2D02
DTSTART;TZID=Europe/Stockholm:20250617T160000
DTEND;TZID=Europe/Stockholm:20250617T163000
UID:submissions.pasc-conference.org_PASC25_sess129_msa240@linklings.com
SUMMARY:From Reactive Debugging to Proactive Detection: ML for Performance
 -Aware Software Development
DESCRIPTION:Tanzima Islam (Texas State University)\n\nSoftware performance
  evolves over time, yet traditional debugging and profiling remain reactiv
 e, costly, and disconnected from development workflows. Performance drift—
 gradual degradation in execution efficiency due to code modifications—ofte
 n goes undetected until it causes significant slowdowns, forcing late-stag
 e debugging and costly fixes. This talk presents a vision for AI/ML-driven
  proactive performance-drift detection, where models continuously monitor 
 software evolution, identifying inefficiencies before they degrade executi
 on. By combining static analysis (abstract syntax trees) with dynamic insi
 ghts from nightly tests, this framework enables early detection of perform
 ance-impacting changes. Traditional ML approaches require full model retra
 ining whenever code changes or new runtime data become available, making t
 hem impractical for fast-moving development cycles. Few-shot learning elim
 inates this overhead by allowing models to update incrementally with minim
 al new data. Attention-based representation learning further enhances inte
 rpretability by prioritizing performance-critical features, enabling more 
 targeted interventions. This framework supports two key decision-making pr
 ocesses where (1) developers can receive automated feedback on whether a c
 ode change improves or degrades performance, enabling early intervention; 
 (2) the insights can guide hardware configuration choices and runtime para
 meter tuning. This approach can be seamlessly integrated into CI/CD pipeli
 nes to achieve software that not only remains correct but also maintains e
 fficiency as it evolves.\n\nDomain: Computational Methods and Applied Math
 ematics\n\nSession Chair: Florina Ciorba (University of Basel)\n\n
END:VEVENT
END:VCALENDAR