Abstract. In this paper we consider latent variable models and introduce a new U-likelihood concept for estimating the distribution over hidden variables. One can derive an estimat...
JaeMo Sung, Sung Yang Bang, Seungjin Choi, Zoubin ...
Abstract. Discriminative and generative methods provide two distinct approaches to machine learning classification. One advantage of generative approaches is that they naturally mo...
Bin covering takes as input a list of items with sizes in (0 1) and places them into bins of unit demand so as to maximize the number of bins whose demand is satis ed. This is in ...
We formulate a principle for classification with the knowledge of the marginal distribution over the data points (unlabeled data). The principle is cast in terms of Tikhonov styl...
Dataset shift from the training data in a source domain to the data in a target domain poses a great challenge for many statistical learning methods. Most algorithms can be viewed ...