BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium: Uncertainty Quantification and Label Error Detectio
 n for Semantic Segmentation
DTSTART:20221024T161500
DTEND:20221024T173000
DTSTAMP:20260407T112356Z
UID:921a14e40cbd92d34cea9af2ed5a466d1b49592ae31b6e1cc935ac6e
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Matthias Rottmann - University of Wuppertal\nVideo of his 
 talk\n\nAbstract\nCurrently\, deep learning is one of the most powerful to
 ols for the automation of complex tasks\, such as environmental perception
  in automated driving or robotics. However\, the lack of proper assessment
  of the reliability of a given prediction often hinders the successful app
 lication of deep learning in practice. In this talk\, we consider the sema
 ntic segmentation of camera images in the context of street scenes. I intr
 oduce an uncertainty quantification method for predictions of deep neural 
 networks on the level of predicted connected components\, show performance
  results and its application to the detection of label errors in semantic 
 segmentation datasets. Besides that\, we get a glimpse of further applicat
 ions.\n\nBio\nMatthias Rottmann achieved his PhD in applied mathematics in
  2016 at University of Wuppertal (UW)\, Germany. He was postdoc at UW's ap
 plied computer science group from 2016 to 2020. In 2020 he got assigned a 
 tenured lecturer position at UW's stochastics group. He is leading a junio
 r research group with focus on uncertainty in deep learning and is current
 ly visiting researcher at EPFL.
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420 https://epfl.zoom.us/
 j/65962421073
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR