Authors ' Addresses

We would like to thank Floris Geerts and Rainer Gemulla for helpful technical discussions. We thank Radu Curticapean for pointing out Bezout's theorem. Learning the parameters of complex probabilistic-relational models from la-beled training data is a standard technique in machine learning, which has been intensively studied in the subeld of Statistical Relational Learning (SRL), but|so far|this is still an under-investigated topic in the context of Probabilistic Databases (PDBs). In this paper, we focus on learning the probability values of base tuples in a PDB from query answers, the latter of which are represented as labeled lineage formulas. Specically, we consider labels in the form of pairs, each consisting of a Boolean lineage formula and a marginal probability that comes attached to the corresponding query an-swer. The resulting learning problem can be viewed as the inverse problem to condence computations in PDBs: given a set of labeled query answers, learn the probability values of the base tuples, such that the marginal prob