Natural concept generation is critical to statistical interlinguabased speech translation performance. To improve maximumentropy-based concept generation, a set of novel features and algorithms are proposed including features enabling model training on parallel corpora, employment of confidence thresholds and multiple sets of features. The concept generation error rate is reduced by 43%-50% in ...

A method is described to incorporate bilexical preferences between phrase heads, such as selection restrictions, in a MaximumEntropy parser for Dutch. The bilexical preferences are modelled as association rates which are determined on the basis of a very large parsed corpus (about 500M words). We show that the incorporation of such selftrained preferences improves parsing accuracy significantly.

A method is described to incorporate bilexical preferences between phrase heads, such as selection restrictions, in a MaximumEntropy parser for Dutch. The bilexical preferences are modelled as association rates which are determined on the basis of a very large parsed corpus (about 500M words). We show that the incorporation of such selftrained preferences improves parsing accuracy significantly.

A new multi-frequency synthesis algorithm for reconstructing images from multi-frequency VLBI data is proposed. The algorithm is based on a generalized maximumentropy method, and makes it possible to derive an effective spectral correction for images over a broad frequency bandwidth, while simultaneously reconstructing the spectral-index distribution over the source. The results of numerical si...

Bayesian networks are widely accepted as tools for probabilistic modeling. In building Bayesian networks in collaboration with domain experts, the de nition of the graphical structure is usually relatively easy. The assessment of the conditional probability tables (CPT) is often a much more diÆcult task, even when there is a lot of statistical information available as domain knowledge. The prob...

The study of matrices with a displacement structure is mainly concerned with recursions for the so-called generator matrices. The recursion usually involves free parameters, which can be chosen in several ways so as to simplify the resulting algorithm. In this correspondence we present a choice for the parameters that is motivated by a maximumentropy formulation. This choice further motivates t...

In this paper we present a novel discriminative mixture model for statistical machine translation (SMT). We model the feature space with a log-linear combination of multiple mixture components. Each component contains a large set of features trained in a maximumentropy framework. All features within the same mixture component are tied and share the same mixture weights, where the mixture weight...

The well known maximum-entropy principle due to Jaynes, which states that given mean parameters, the maximum entropy distribution matching them is in an exponential family has been very popular in machine learning due to its “Occam’s razor” interpretation. Unfortunately, calculating the potentials in the maximumentropy distribution is intractable [BGS14]. We provide computationally efficient ve...

In this paper we compare and contrast two approaches to Machine Translation (MT): the CMU-UKA Syntax Augmented Machine Translation system (SAMT) and UPC-TALP N-gram-based Statistical Machine Translation (SMT). SAMT is a hierarchical syntax-driven translation system underlain by a phrase-based model and a target part parse tree. In N-gram-based SMT, the translation process is based on bilingual ...

Moment closures of gaskinetic theory offer a method for the prediction of non-equilibrium flows which have a wider range of validity than standard continuum methods, such as solution of the Navier-Stokes equations, and can be much more computationally efficient than particle-based techniques, such as Direct Simulation Monte Carlo (DSMC). Moment methods yield systems of first-order partial diffe...