We study the sparsity of spectro-temporal representation of speech in reverberant acoustic conditions. This study motivates the use of structured sparsity models for efficient speech recovery. We formulate the underdetermined convolutive speech separation in spectro-temporal domain as the sparse signal recovery where we leverage model-based recovery algorithms. To tackle the ambiguity of the re...

Over the last few years, the development of multi-channel sensors motivated interest in methods for the coherent processing of multivariate data. Some specific issues have already been addressed as testified by the wide literature on the so-called blind source separation (BSS) problem. In this context, as clearly emphasized by previous work, it is fundamental that the sources to be retrieved pr...

Predicting users political party in social media has important impacts on many real world applications such as targeted advertising, recommendation and personalization. Several political research studies on it indicate that political parties’ ideological beliefs on sociopolitical issues may influence the users political leaning. In our work, we exploit users’ ideological stances on controversia...

For high-dimensional problems various parametric priors have been proposed to promote sparse solutions. While parametric priors has shown considerable success they are not very robust in adapting to varying degrees of sparsity. In this work we propose a discrete mixture prior which is partially nonparametric. The right structure for the prior and the amount of sparsity is estimated directly fro...

In Data Warehouse environments data sparsity is a common issue. In the past few years a number of different techniques have been proposed to tackle this issue, but it remains an unresolved problem. The majority of the techniques developed to date to tackle this problem have been based on the relational model. At the same time other research has focused on alternative data models that abandon th...

We address the task of annotating images with semantic tuples. Solving this problem requires an algorithm able to deal with hundreds of classes for each argument of the tuple. In such contexts, data sparsity becomes a key challenge. We propose handling this sparsity by incorporating feature representations of both the inputs (images) and outputs (argument classes) into a factorized log-linear m...

For distributed optical fiber pipeline pre-warning system, the sampling rate used is very high and thus huge data will be generated, which makes it difficult to transfer and store. Compressive sensing is a new compressed sampling method in the field of signal processing which compresses and samples the signal simultaneously. In this paper, an adaptive compressive sensing method is presented for...

l1-SPIRiT is a fast magnetic resonance imaging (MRI) method which combines parallel imaging (PI) with compressed sensing (CS) by performing a joint l1-norm and l2-norm optimization procedure. The original l1-SPIRiT method uses two-dimensional (2D) Wavelet transform to exploit the intra-coil data redundancies and a joint sparsity model to exploit the inter-coil data redundancies. In this work, w...

Traditional dictionary learning method work by vectorizing long signals, and training on the frames of the data, thereby restricting the learning to time-localized atoms. We study a shift-tolerant approach to learning dictionaries, whereby the features are learned by training on shifted versions of the signal of interest. We propose an optimized Subspace Clustering learning method to accommodat...

Traditional tensor decompositions such as the CANDECOMP / PARAFAC (CP) and Tucker decompositions yield higher-order principal components that have been used to understand tensor data in areas such as neuroimaging, microscopy, chemometrics, and remote sensing. Sparsity in high-dimensional matrix factorizations and principal components has been well-studied exhibiting many benefits; less attentio...