Variational methods for approximate inference in machine learning often adapt a parametric probability distribution to optimize a given objective function. This view is especially ...
Antti Honkela, Matti Tornio, Tapani Raiko, Juha Ka...
We introduce relational grams (r-grams). They upgrade n-grams for modeling relational sequences of atoms. As n-grams, r-grams are based on smoothed n-th order Markov chains. Smoot...
: To manage information quality (IQ) effectively, one needs to know how IQ changes over time, what causes it to change, and whether the changes can be predicted. In this paper we a...
Transductive inference on graphs such as label propagation algorithms is receiving a lot of attention. In this paper, we address a label propagation problem on multiple networks a...
We give a new class of outer bounds on the marginal polytope, and propose a cutting-plane algorithm for efficiently optimizing over these constraints. When combined with a concav...