Jeudi 17 mars, 11h, salle 3.124 bat 5, ou lien visio : https://umontpellier-fr.zoom.us/my/carral
Title: The Complexity Landscape of Probabilistic Query Evaluation
By: Ismail Ilkan Ceylan (U. Oxford)
Abstract: Large-scale probabilistic knowledge bases are becoming increasingly important in academia and industry. They are continuously extended with new data, powered by modern information extraction tools. Probabilistic databases are data models which provide means for representing, storing, and processing large-scale probabilistic data. In this talk, we will introduce probabilistic databases, establish the relation between probabilistic query evaluation and the task of weighted model counting, and provide an overview of known complexity results. We will then discuss probabilistic query evaluation relative to more elaborate query languages (i.e., ontology-mediated queries) and show that probabilistic query evaluation is inherently harder for such class queries. Going beyond exact inference, we will briefly discuss the approximability status of probabilistic query answering for different query languages.
Bio: Ismail Ilkan Ceylan obtained his PhD at TU Dresden, Germany, with this thesis entitled “Query answering over probabilistic data and knowledge bases” which was awarded the Beth Dissertation Prize. Ismail is currently working as a lecturer at the Department of Computer Science, University of Oxford. Previously, he was working as a postdoctoral researcher at the University Oxford. Ismail’s research interests are broadly in AI and machine learning with a particular focus on relational learning and reasoning. The goal is being able to more efficiently and reliably learn from relational patterns and reason over them. This is a highly interactive field, where techniques from machine learning, knowledge representation, and theoretical computer science are relevant.
