BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.pydata.org//pydata-eindhoven-2025//talk//GJNEKL
BEGIN:VEVENT
UID:pretalx-pydata-eindhoven-2025-GJNEKL@cfp.pydata.org
DTSTART:20251209T141000Z
DTEND:20251209T144000Z
DESCRIPTION:Large AI models have become powerful but increasingly impractic
 al\; with escalating training costs\, bloated memory requirements\, and la
 tency bottlenecks that limit real-world deployments. This talk introduces 
 CompactifAI: a quantum-inspired compression framework that uses tensor net
 works to surgically shrink large models while preserving their accuracy an
 d capabilities.
DTSTAMP:20260425T204117Z
LOCATION:Planck-Bohr
SUMMARY:CompactifAI: Quantum-Inspired AI Model Compression - Jon Leiñena O
 tamendi
URL:https://cfp.pydata.org/pydata-eindhoven-2025/talk/GJNEKL/
END:VEVENT
END:VCALENDAR