Models 1. 5-Nearest Neighbor with Dynamic Time Warping (DTW)? 2. 5-Nearest Neighbor with Complexity-Invariant Distance (CID)? 3. Regression on features (Reg) Method 1. 10 fold cross validation 2. Lowest quality training cases duplicated 3. Bayesian approach using Gibbs sampling 4. Model from best algorithm used to classify 90,631 cases 5. Labels correlated with 30-day outcomes (4) Times Series ...

A primary goal of the 1996 Telecommunications Act is to encourage competition in longdistance telephone markets. Four years after passage of this legislation, Bell Operating Companies (“BOCs”) have been granted permission to offer long-distance services in only one state. The regulatory barrier to entry is justified on grounds that the BOCs have the ability to discriminate against incumbent lon...

We propose a parallel ray-tracing technique to visualize signed distance fields generated from triangular meshes based on NVIDIA OptiX. Our method visualizes signed distance fields with various distance offset values at interactive rates (2–12 fps). Our method utilizes a parallel kd-tree implementation to query the nearest triangle and the sphere tracing method to visualize the implicit surface.

This paper proposes a novel pattern classiication approach, called the nearest linear combination (NLC) approach, for eigenface based face recognition. Assume that multiple prototypical vectors are available per class, each vector being a point in an eigenface space. A linear combination of prototypical vectors belonging to a face class is used to deene a measure of distance from the query vect...

We suggest a simple modification to the kd-tree search algorithm for nearest neighbor search resulting in an improved performance. The Kd-tree data structure seems to work well in finding nearest neighbors in low dimensions but its performance degrades even if the number of dimensions increases to more than three. Since the exact nearest neighbor search problem suffers from the curse of dimensi...

Nonsmooth analysis of a broad class of functions taking the form F (x) = mini fi(x), where each fi is a convex function. One element of this class of functions is the distance function, which measures the distance between a point and the nearest point on the nearest obstacle. Many motion planning algorithms are based on the distance function, and thus rigorous analysis of the distance function ...

We suggest a simple modification to the Kd-tree search algorithm for nearest neighbor search resulting in an improved performance. The Kd-tree data structure seems to work well in finding nearest neighbors in low dimensions but its performance degrades even if the number of dimensions increases to more than two. Since the exact nearest neighbor search problem suffers from the curse of dimension...