We introduce the Concave-Convex procedure (CCCP) which constructs discrete time iterative dynamical systems which are guaranteed to monotonically decrease global optimization/ener...
Marching Ridges is an algorithm for finding ridges of measurement functions defined for images, in which the ridges are defined as level sets of first derivatives of the measureme...
We propose a directed graphical representation of utility functions, called UCP-networks, that combines aspects of two existing preference models: generalized additive models and ...
Within the task of collaborative filtering two challenges for computing conditional probabilities exist. First, the amount of training data available is typically sparse with resp...
In this paper we introduce a new underlying probabilistic model for principal component analysis (PCA). Our formulation interprets PCA as a particular Gaussian process prior on a ...