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...
Continuously-Adaptive Discretization for Message-Passing (CAD-MP) is a new message-passing algorithm for approximate inference. Most message-passing algorithms approximate continu...
A novel random text generation model is introduced. Unlike in previous random text models, that mainly aim at producing a Zipfian distribution of word frequencies, our model also ...
Although kernel measures of independence have been widely applied in machine learning (notably in kernel ICA), there is as yet no method to determine whether they have detected st...
Arthur Gretton, Kenji Fukumizu, Choon Hui Teo, Le ...
— The amount of data stored in data warehouses grows very quickly so that they can get saturated. To overcome this problem, we propose a language for specifying forgetting functi...