About the project
Inconsistencies often occur in the analysis of data, in the form of contradictions between various parts of the data, between parts of the background knowledge which enters models, between predictions/estimates and what is known about these in advance, or with respect to a combination thereof. This project aims at bringing together the views and methods of statistics and logic on detecting and handling inconsistencies. The PhD student must expect to do exciting interdisciplinary work between these two scientific fields. The supervising team has expertise from both.
The project has two specific aims: (i) develop probabilistic models for partially observed preference data (e.g., but not limited to, user clicking data) for preference learning in the presence of inconsistencies in the input data; (ii) develop probabilistic approaches to the treatment of inconsistencies beyond prefernce learning, in more general settings. It will be useful to develop a suitable formal language to describe preference statements that enables detection and inference in the presence of inconsistencies and to handle inconsistencies as a piece of useful information.
Project team
- Associate Professor Ida Scheel, University of Oslo, Department of Mathematics
- Professor Martin Giese, University of Oslo, Department of Informatics
- Associate Professor Valeria Vitelli, University of Oslo, Department of Informatics
- Professor Arnoldo Frigessi, University of Oslo
- Associate Professor Anne-Marie George, University of Oslo, Department of Informatics
- Phd Fellow Lorenzo Zuccato, University of Oslo, Department of Mathematics