Francesco Giannini (DIISM, University of Siena)
Oct 24, 2018 – 11:00 AM
DIISM, Artificial Intelligence laboratory (room 201), Siena SI
While learning from constraints is a main topic in artificial intelligence, the problem of learning constraints from exam-
ples has received less attention. In this talk, we focus on constraints expressed by logical formulas and some approaches considered by different authors in literature are discussed. Another relevant topic for learning of constraints is how to learn the confidence associated to the inducted rules. In fact, some relational knowledge about the data can hold partially. Finally, some advance in learning from constraints jointly with the weigth learning associated to the rules is presented.