BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Human Values and Explainable Artificial Intelligence - Dr. Rune Ny
 rup\, Center for the Future of Intelligence
DTSTART:20190429T110000Z
DTEND:20190429T123000Z
UID:TALK114133@talks.cam.ac.uk
CONTACT:Tellef S. Raabe
DESCRIPTION:A common objection to the use of artificial intelligence in de
 cision-making is the concern that it is often difficult to explain or unde
 rstand how AI systems make decisions. There is a growing body of technical
  AI research developing techniques for making AI more “explainable” or
  “interpretable”. However\, it is still not well understood why this i
 s an important property for an AI system to possess\, or what types of exp
 lanations are most important. While there are empirical studies of which t
 ypes of explanations individuals subjected to AI decision-making find sati
 sfactory\, psychological evidence suggests people’s sense of understandi
 ng is often unreliable and easy to manipulate. In this paper\, I argue tha
 t a pragmatist account of explanation provides a fruitful framework for ex
 ploring the problem of AI Explainability\, which allows us to combine norm
 ative and empirical perspectives on user values.
LOCATION:17 Mill Lane\, Room B (1st floor)
END:VEVENT
END:VCALENDAR