Speaker: Tarek R. Besold
Research Centre: Centre for Machine Learning
This talk will provide an introduction to (some of the) current discussions surrounding questions of explainability in AI and ML. On the technical side, we will characterize four notions of explainable AI/ML that cut across research fields: opaque systems that offer no insight into their algorithmic mechanisms; interpretable systems where users can mathematically analyse their algorithmic mechanisms; comprehensible systems that emit symbols enabling user-driven explanations of how a conclusion is reached; and explainable systems, where automated reasoning is central to output crafted explanations without requiring human post processing as final step of the generative process. On the conceptual side, we will look into current societal and regulatory efforts trying to address the desire and need for explainability in an AI/ML context.
Tarek is a Lecturer in Data Science at City, University of London, conducting research at the intersection between artificial intelligence, computational creativity, and cognitive systems. He studied mathematics, computer science, and logic in Erlangen, Zaragoza, and Amsterdam, obtained a PhD from the Institute of Cognitive Science in Osnabrück, and worked as a postdoctoral researcher at the KRDB Research Centre of the Free University of Bozen-Bolzano and at the Digital Media Lab of the University of Bremen. Among others, Tarek was the General Chair of the HLAI 2016 Joint Multi-Conference on Human-Level Artificial Intelligence, and founder and/or organizer of several international workshop series bridging between AI and cognitive science. He was co-editor of the books “Computational Creativity Research: Towards Creative Machines” and “Concept Invention: Foundations, Implementation, Social Aspects and Applications”, and serves in different editorial functions for the Journal of Artificial Intelligence Research (JAIR), for Frontiers in Psychology: Cognition, and for Biologically Inspired Cognitive Architectures (BICA).
Slides from this seminar can be found here.
Share this event
When and where
6.00pm - 7.00pmWednesday 22nd November 2017