The following is a list of the major changes since Version 2.0 of the package. Replaced gsvd by cgsvd which has a diierent sequence of output arguments. Removed the obsolete function csdecomp (which replaced the function csd) Deleted the function mgs. Changed the storage format of bidiagonal matrices to sparse, instead of a dense matrix with two columns. Removed the obsolete function bsvd. Adde...

A multitask learning framework is developed for discriminative classification and regression where multiple large-margin linear classifiers are estimated for different prediction problems. These classifiers operate in a common input space but are coupled as they recover an unknown shared representation. A maximum entropy discrimination (MED) framework is used to derive the multitask algorithm w...

INTRODUCTION Electrical impedance tomography (EIT) is an emerging clinical tool for monitoring ventilation distribution in mechanically ventilated patients, for which many image reconstruction algorithms have been suggested. We propose an experimental framework to assess such algorithms with respect to their ability to correctly represent well-defined physiological changes. We defined a set of ...

In this paper we propose the sparse supervised topic model (SSTM), a graphical model that learns topic structures of a given document collection and also a sparse linear prediction model for response vairables associated with documents. Our model jointly learns the topics and the classifier and encourages a sparse classifier by concentrating all the relevant information for prediction into a sm...

EEcient algorithm for model further extended to allow \dynamic membership" (to appear at INFOCOM'98). EEcient algorithm for low-degree multicast tree problem. 19 Conclusions Proposed algorithm produces near-optimal multicast trees for small and sparse networks. Quality of solution is maintained even when size and density of networks grow. Computation time is polynomially bounded and expected to...

Truncated singular value decomposition is a reduced version of the in which only few largest values are retained. This paper presents novel perturbation analysis for truncated real matrices. First, we describe expansions truncation order r . We extend results subspace to derive first-order expansion operator about matrix with rank greater than or equal Observing that can be greatly simplified w...

I introduce two strategies to overcome the slow convergence of least squares sparse data interpolation: 1) a 2-D multiscale Laplacian regularization operator, and 2) an explicit quadtree-style upsampling scheme which produces a good initial guess for iterative schemes. The multiscale regularization produces an order-of-magnitude speedup in the interpolation of a sparsely sampled topographical m...

  In this paper, modeling and optimization of Fischer-Tropsch Synthesis is considered in a fixed-bed catalytic reactor using an industrial Fe-Cu-K catalyst. A one dimensional pseudo-homogenous plug flow model without axial dispersion is developed for converting syngas to heavy hydrocarbons. The effects of temperature, pressure, H2 to CO ratio in feed stream, and CO molar flow on the mass flow r...

In this paper, it was an attempt to be present a practical perishability inventory model. The proposed model adds using spoilage of products and variable prices within a time period to a recently published location-inventory-routing model in order to make it more realistic. Aforementioned model by integration of strategic, tactical and operational level decisions produces better results for sup...

Texts are the major information carrier for internet users, from which learning latent representations has important research and practical value. Neural topic models have been proposed great performance in extracting interpretable topics of texts. However, there remain two limitations: (1) these methods generally ignore contextual texts limited feature representation ability due to shallow fee...