Publikation

Recent Trends in XAI: A Broad Overview on current Approaches, Methodologies and Interactions

Jakob M. Schönborn, Klaus-Dieter Althoff

In: Workshops Proceedings for the Twenty-seventh International Conference on Case-Based Reasoning: Explainable Knowledge in Computational Design, Media, and Teaching workshop. Workshop on Explainable Knowledge in Computational Design, Media, and Teaching (EK-CDMT-2019) befindet sich Twenty-seventh International Conference on Case-Based Reasoning September 8-12 Otzenhausen Germany CEUR-WS 2019.

Abstrakt

The definition of an explainable artificial intelligence heavily depends on the use-case, whether one is focusing on the technical knowledge-management component [30, 33, 37, 43] or rather the more social interaction including speech acts and conversations [27, 31, 33]. Since the uprising debate of the unknown outcome on the development of AI in general using Deep Learning [4, 34, 35, 44] and recent legal restrictions (for example the GDPR [19]), the need on developing an explainable AI is rapidly increasing, especially since the last two years. Additionally, the goal to increase the users trust towards AI has still to be achieved. Thus, this contribution aims to provide an overview on the current topics especially since 2018 with a focus on case-based explanations up until today.

2019-Recent_Trends_in_XAI_A_Broad_Overview_on.pdf (pdf, 1 MB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence