BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.4.0.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:375@isdm.umontpellier.fr
DTSTART;TZID=Europe/Paris:20240402T140000
DTEND;TZID=Europe/Paris:20240402T140000
DTSTAMP:20260828T111905Z
URL:https://isdm.umontpellier.fr/events/runtime-monitoring-of-deep-neural-
 networks-2/
SUMMARY:Runtime Monitoring of Deep Neural Networks
DESCRIPTION:Room 02.124\, Building 5\, St Priest campus\n\nMachine Learning
  in Montpellier\, Theory &amp\; Practice\n\nThe use of deep neural network
 s in critical domains\, such as aviation\, is limited by our ability to en
 sure their correct behavior. Execution monitors are components aimed at id
 entifying dangerous predictions and discarding them before they lead to ca
 tastrophic consequences. Several recent works on real-time monitoring have
  focused on detecting out-of-distribution (OOD) inputs\, i.e.\, identifyin
 g inputs that differ from the training data. In this presentation\, we wil
 l show that OOD detection is not a well-suited framework for designing eff
 ective execution monitors and that it is more relevant to evaluate monitor
 s based on their ability to discard incorrect predictions. We call this pa
 radigm out-of-model-scope (OMS) detection and discuss the conceptual diffe
 rences with OOD. We will also present in-depth experiments to show that st
 udying monitors in the OOD framework can be misleading: 1. very good OOD r
 esults can give a false sense of security\, 2. an OOD-based comparison may
  not identify the best monitor for detecting errors.\n\nMachine Learning i
 n Montpellier\, Theory &amp\; Practice
ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/
 2026/06/ml-mtp-gC78d5.png
CATEGORIES:ML MTP
LOCATION:Saint Priest Campus - Building 5 - Room 02.124\, 860 rue St Priest
 \, Montpellier\, 
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=860 rue St Priest\, Montpel
 lier\, ;X-APPLE-RADIUS=100;X-TITLE=Saint Priest Campus - Building 5 - Room
  02.124:geo:0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:DAYLIGHT
DTSTART:20240331T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR