Interpretation of x-ray absorption near-edge structure (XANES) experiments is often done via analyzing the role of particular atoms in the formation of specific peaks in the calculated spectrum. Typically, this is achieved by calculating the spectrum for a series of trial structures where various atoms are moved and/or removed. A more quantitative approach is presented here, based on comparing ...

We consider the problem of learning a hypergraph using edge-detecting queries. In this model, the learner may query whether a set of vertices induces an edge of the hidden hypergraph or not. We show that an r-uniform hypergraph with m edges and n vertices is learnable with O(24rm · poly(r, logn)) queries with high probability. The queries can be made in O(min(2r(logm + r)2, (logm + r)3)) rounds...

This paper presents an integrated approach for robustly locating facial landmark for drivers. In the first step a cascade of probability learners is used to detect the face edge primitives from fine to coarse, so that faces with variant head poses can be located. The edge density descriptors and skin-tone color features are combined together as the basic features to examine the probability of a...

We determine to within a constant factor the threshold for the property that two random k-uniform hypergraphs with edge probability p have an edge-disjoint packing into the same vertex set. More generally, we allow the hypergraphs to have different densities. In the graph case, we prove a stronger result, on packing a random graph with a fixed graph.

A method for labelling edge data in registered intensity and range images is presented. First, depth, orientation and intensity discontinuities are identified by one-dimensional filters aligned with the image axes. Second, the filtered outputs are combined by a maximum a-posteriori probability estimation to produce a set of edge labels at each site within the dual lattice.

A few edge detectors are derived from the contrast ratio edge detector to extract linear features from SAR imagery with a constant probability of false alarm. But all of these detectors need one or more thresholds, which are generally predefined or selected by experience, to determine that the pixel belongs to an edge or not. In this paper, we propose a method to automatically determine the thr...

We propose graphical sure screening, or GRASS, a very simple and computationally-efficient screening procedure for recovering the structure of a Gaussian graphical model in the high-dimensional setting. The GRASS estimate of the conditional dependence graph is obtained by thresholding the elements of the sample covariance matrix. The proposed approach possesses the sure screening property: with...

The integration of spatial context in the classification of hyperspectral images is known to be an effective way in improving classification accuracy. In this paper, a novel spectralspatial classification framework based on edge-preserving filtering is proposed. The proposed framework consists of the following three steps. First, the hyper-spectral image is classified using a pixel-wise classif...

We study spectral clustering techniques to learn community structures in labeled random graphs where edge labels from a label set L = {1, ..., L} are drawn according to discrete probability distributions parametrized by community membership of the two end-nodes of the edge. This is a strict generalization of the standard stochastic block model for community detection.

Let $G$ be a connected graph with minimum degree $delta$ and edge-connectivity $lambda$. A graph ismaximally edge-connected if $lambda=delta$, and it is super-edge-connected if every minimum edge-cut istrivial; that is, if every minimum edge-cut consists of edges incident with a vertex of minimum degree.In this paper, we show that a connected graph or a connected triangle-free graph is maximall...