BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Explainable AI for image classification models: saliency maps and 
 interpretable classification
DTSTART:20251125T101500Z
DTEND:20251125T111500Z
DTSTAMP:20260425T102800Z
UID:indico-event-14457@indico.math.cnrs.fr
DESCRIPTION:Speakers: Ronan Sicre (IRIT)\n\nWe will first review previous 
 work on explainable AI methods and present two works on this topic. We wil
 l first present Opti-CAM\, a CAM-based method that optimizes a masking obj
 ective per instance. The resulting saliency map highlights the important a
 rea of an image regarding the decision of a trained image classification n
 etwork. Then\, we will focus on image classification using parts or protot
 ype-based architectures. Such model is explainable by design and brings tr
 ansparency to a model.\n\nhttps://indico.math.cnrs.fr/event/14457/
LOCATION:Salle K. Johnson (1R3\, 1er étage)
URL:https://indico.math.cnrs.fr/event/14457/
END:VEVENT
END:VCALENDAR