BEGIN:VCALENDAR
VERSION:2.0
PRODID:https://github.com/derhansen/sf_event_mgt
METHOD:PUBLISH
BEGIN:VEVENT
UID:3791-2675@rg-trier-luxemburg.gi.de
CLASS: PUBLIC
SUMMARY:AI4Health Lecture Series: Ute Schmid and Bettina Finzel
DESCRIPTION:The next talk of the AI4Health lecture series will continue on 
 November 04 with a talk on "Learning from Mutual Explanations for Cooperati
 ve Decision Making in Medicine" by Ute Schmid and Bettina Finzel (Bamberg U
 niversity, Germany).\n\nLearning from Mutual Explanations for Cooperative D
 ecision Making in Medicine\n\nAbstract: Medical decision making is one of t
 he most relevant real world domains where intelligent support is necessary 
 to help human experts master the ever growing complexity. At the same time,
  standard approaches of data driven black box machine learning are not reco
 mmendable since medicine is a highly sensitive domain where errors may have
  fatal consequences. In the talk, we will advocate interactive machine lear
 ning from mutual explanations to overcome typical problems of purely data d
 riven approaches to machine learning. Mutual explanations, realised with th
 e help of an interpretable machine learning approach, allow to incorporate 
 expert knowledge in the learning process and support the correction of erro
 neous labels as well as dealing with noise. Mutual explanations therefore c
 onstitute a framework for explainable, comprehensible and correctable class
 ification. Specifically, we present an extension of the inductive logic pro
 gramming system Aleph which allows for interactive learning. We introduce o
 ur application LearnWithME which is based on this extension. LearnWithME ge
 ts input from a classifier such as a Convolutional Neural Net‘s prediction 
 on medical images. Medical experts can ask for verbal explanations in order
  to evaluate the prediction. Through interaction with the verbal statements
  they can correct classification decisions and in addition can also correct
  the explanations. Thereby, expert knowledge is taken into account in form 
 of constraints for model adaptation.\n\n- See more at: https://acc.uni.lu/a
 i4health\n\n\n\nWe invite you to join this Webinars.\n\nMeeting link: https
 ://unilu.webex.com/unilu/j.php?MTID=m060774148c89025957b2831d3031c1fc\n\nMe
 eting number (access code): 163 267 6786 Meeting password: UL-AIForHCWebs\n
 \nHost key: 478625 \n\nSee more at: acc.uni.lu/ai4health\n\n\n\n
LOCATION:Online
DTSTAMP:20200928T095004Z
DTSTART:20201104T150000Z
DTEND:20201104T170000Z
END:VEVENT
END:VCALENDAR