CIN Dialogue 2016: Wird es maschinelles Bewusstsein geben? Ein interdisziplinäres Streitgespräch
Discussants are computer scientist and machine learning specialist Joachim M. Buhmann (Eidgenössische Techniche Hochschule Zürich) and neurophysiologist Andreas K. Engel (Universitätsklinikum Hamburg-Eppendorf). The discussion will be chaired and moderate
December 01 2016, 19h c.t.
Lecture Hall Audimax, Neue Aula, Geschwister-Scholl-Platz, Uni Tübingen
Is the brain a kind of `biological machine‘? Can computers fulfil its functions? And where is the boundary between artificial intelligence and artificial consciousness? These are fundamental questions that can only be approached on an interdisciplinary footing.
Programming machines traditionally happens ‘top-down’: programmers strive to provide the machine with fitting reactions to every possible input. Such a top-down ruleset is not truly intelligent. It can only fulfil specific tasks which are established beforehand. For a long time, computer science has endeavoured to develop truly intelligent machines capable of (‘bottom-up’) learning. But only for a few years now has this new wave of artificial intelligence been influencing our everyday lives.
Algorithms of self-directed learning are modelled on human learning processes. They have brought us Siri and a computer that can beat a Go master at their own game. Vehicles driving themselves will soon be a reality. But could machines also learn to reflect their own existence? Would they not need a consciousness for that?
This touches on a much more basic question: What exactly do we consider consciousness in machines to be? To answer it, we would need to explain and understand consciousness in scientifically rigorous and empirically founded terms. Our human consciousness arises in the brain, from cell architectures and electrical impulses. Can we analyse and understand our biologically founded consciousness to such a degree that we can use it as a model for constructing machines with similar networking?
This year’s CIN Dialogue focuses on a topic that might seem to be fully within the purview of computer science at first glance. But today more than ever before, it can profit from the neuroscientific perspective. Discussants are computer scientist and machine learning specialist Joachim M. Buhmann (Eidgenössische Techniche Hochschule Zürich) and neurophysiologist Andreas K. Engel (Universitätsklinikum Hamburg-Eppendorf). The discussion will be chaired and moderated by Ulrich Schnabel (DIE ZEIT).
Learn more about CIN Dialogues here.
- Werner Reichardt Centre for Integrative Neuroscience
- University of Tübingen