Recently popularized randomized methods for principal component analysis (PCA) efficiently and reliably produce nearly optimal accuracy -- even on parallel processors -- unlike the...
This paper develops an efficient new method for 3D partial shape retrieval. First, a Monte Carlo sampling strategy is employed to extract local shape signatures from each 3D model...
We analyze the technique for reducing the complexity of entropy coding consisting in the a priori grouping of the source alphabet symbols, and in dividing the coding process in tw...
Discovering accurate and interesting classification rules is a significant task in the post-processing stage of a data mining (DM) process. Therefore, an optimization problem exis...
We present a class of graphical models for directly representing the joint cumulative distribution function (CDF) of many random variables, called cumulative distribution networks...