We present a polytope-kernel density estimation (PKDE) methodology that allows us to perform exact mean-shift updates along the edges of the Delaunay graph of the data. We discuss...
Abstract— Research on numerical solution methods for partially observable Markov decision processes (POMDPs) has primarily focused on discrete-state models, and these algorithms ...
We introduce a generic and efficient method for 2D and 3D shape estimation via density Þelds. Our method models shape as a density map and uses the notion of density to Þt a mod...
In this paper, we present a Question Answering system called KUQA (Korea University Question Answering system) developed by using semantic categories and co-occurrence density. Se...
Soo-Min Kim, Dae-Ho Baek, Sang-Beom Kim, Hae-Chang...
A central object of study in the field of algorithmic randomness are notions of randomness for sequences, i.e., infinite sequences of zeros and ones. These notions are usually def...