This paper develops a general, formal framework for modeling term dependencies via Markov random fields. The model allows for arbitrary text features to be incorporated as eviden...
Signal modeling lies at the core of numerous signal and image processing applications. A recent approach that has drawn considerable attention is sparse representation modeling, in...
We present three approaches for unsupervised grammar induction that are sensitive to data complexity and apply them to Klein and Manning's Dependency Model with Valence. The ...
Valentin I. Spitkovsky, Hiyan Alshawi, Daniel Jura...
Many real problems can be modelled as robust shortest path problems on digraphs with interval costs, where intervals represent uncertainty about real costs and a robust path is not...
The Gaussian mixture model (GMM) can approximate arbitrary probability distributions, which makes it a powerful tool for feature representation and classification. However, it su...