Pedro Cabalar: Explaining a Representation Requires Representing Explanations
Pedro Cabalar talks about explainable AI and how we need an effort in knowledge representation to achieve explainability.
TU Wien, Campus Gußhaus
EI 8 Pötzl-Hörsaal
1040 Vienna, Gußhausstraße 27-29
Stiege 1, Erdgeschoß, Raum CDEG08
Explaining a Representation requires Representing Explanations
Achieving a truly Explainable Artificial Intelligence is nowadays a challenge not only for machine learning “black box” techniques but even for symbolic automated reasoning, especially when the explanations of a conclusion are too large or hard to follow for an average human user. In the talk, I will discuss that, to obtain human-oriented explanations, we actually need an additional effort of knowledge representation. In other words, to explain a result or conclusion, we must also specify the form of the explanations we want to obtain and their relation to the result to be explained. I will illustrate some of these ideas with a pair of practical examples for the tool xclingo, that allows incorporating explanation annotations to logic programs under the Answer Set Programming paradigm.
About Pedro Cabalar
Pedro Cabalar is an associate professor at the Computer Science and Information Technology Department, University of A Coruña, Spain. He earned his Bachelor’s degree at the University of Santiago de Compostela and graduated in Computer Science from the Polytechnic University of Madrid, receiving his PhD from the University of A Coruña in 2001. His research interests are focused on Artificial Intelligence and Knowledge Representation and Reasoning with special emphasis on temporal, causal, epistemic and qualitative spatial reasoning, mostly using non-classical logics and logic programming. Many of his contributions are related to logical foundations and extensions of Answer Set Programming. He is also interested in logic based explanations of machine learning classifiers. Pedro Cabalar has collaborated with more than twenty coauthors from universities of nine different countries, publishing in relevant journals like AIJ, TPLP, AMAI, AI Communications, the Journal of Applied Non-Classical Logics or the Logic Journal of the IGPL, and reviewing for other journals like JAIR, J. of Logic and Computation, J. of Applied Logic or ACM TOCL. He is standard editor of the Artificial Intelligence Journal, and member of the advisory board of Theory and Practice of Logic Programming. He has regularly published in the main relevant international conferences like IJCAI, ECAI, KR, AAAI, JELIA, ICLP and LPNMR, being a frequent PC member for these events and acting as senior/area PC for IJCAI, ECAI and KR. He has also co-chaired and organised the 12th edition of LPNMR in 2013 in A Coruña. Pedro Cabalar is the general coordinator of the Master in Artificial Intelligence offered by the three public Galician Universities. In the past he has served as area editor for the Association of Logic Programming Newsletter, he has been scientific editor of the Iberoamerican Journal on Artificial Intelligence and coordinated the PhD and MSc programmes of his Department. Pedro Cabalar has been principal investigator of four national research projects related to Knowledge Representation and Automated Reasoning with applications to the medical domain. He is currently a workgroup leader in the European COST Action DigForASP, on Automated Reasoning applied to Digital Forensics.
About Current Trends in Computer Science
If you are studying with us, the lecture series can be credited as an elective course for students of master programs of computer science: 195.072 Current Trends in Computer Science. Additionally, you can join courses held by this year’s guest professors of our doctoral colleges and the TU Wien Informatics Doctoral School.