Sharif Digital Repository

We apply the theory of optimal transport to study mathematical properties of mean field variational Bayesian approximation. It turns out that if K +C > 0 where C is the convexity coefficient of −log p and K is a lower bound for the Ricci curvature of the underlying parameter space, then the corresponding system of equations of variational Bayesian approximation admits a unique solution. The uniqueness property in presence of symmetry leads to preservation of mode. As an explicit application we correct Bayesian Gaussian Mixture model in such a way that it turns into a convex model while its (unique) maximum likelihood solution coincides asymptotically with the true solution. Using convexity... 

The long-term performance of LiFi networks significantly depends on the location of access points. The optimized placement can be determined based on the distribution of users in the room. In this paper, we investigate the placement optimization for average throughput maximization in the presence of asymmetric distributions. In particular, we represent users' distribution in the indoor environment by the Gaussian mixture model, which is powerful and computationally convenient. Then we obtain the optimized deployment for different scenarios using gradient ascent algorithm. The results show that optimization of deployment significantly improves the average throughput of the network. As the... 

Monte Carlo Approaches for calculating Value-at-Risk (VaR) are powerful tools widely used by financial risk managers across the globe. However, they are time consuming and sometimes inaccurate. In this paper, a fast and accurate Monte Carlo algorithm for calculating VaR and ES based on Gaussian Mixture Models is introduced. Gaussian Mixture Models are able to cluster input data with respect to market's conditions and therefore no correlation matrices are needed for risk computation. Sampling from each cluster with respect to their weights and then calculating the volatility-adjusted stock returns leads to possible scenarios for prices of assets. Our results on a sample of US stocks show that... 

Cardiovascular Disease (CVD) is considered as one of the principal causes of death in the world. Over recent years, this field of study has attracted researchers' attention to investigate heart sounds' patterns for disease diagnostics. In this study, an approach is proposed for normal/abnormal heart sound classification on the Physionet challenge 2016 dataset. For the first time, a fixed length feature vector; called i-vector; is extracted from each heart sound using Mel Frequency Cepstral Coefficient (MFCC) features. Afterwards, Principal Component Analysis (PCA) transform and Variational Autoencoder (VAE) are applied on the i-vector to achieve dimension reduction. Eventually, the reduced... 

The low-dimensional i-vector representation of speech segments is used in the state-of-the-art text-independent speaker verification systems. However, i-vectors were deemed unsuitable for the text-dependent task, where simpler and older speaker recognition approaches were found more effective. In this work, we propose a straightforward hidden Markov model (HMM) based extension of the i-vector approach, which allows i-vectors to be successfully applied to text-dependent speaker verification. In our approach, the Universal Background Model (UBM) for training phrase-independent i-vector extractor is based on a set of monophone HMMs instead of the standard Gaussian Mixture Model (GMM). To... 

Most current speech recognition systems use Hidden Markov Models (HMMs) to deal with the temporal variability of speech and Gaussian mixture models (GMMs) to determine how well each state of each HMM fits a frame or a short window of frames of coefficients that represents the acoustic input. In these systems acoustic inputs are represented by Mel Frequency Cepstral Coefficients temporal spectrogram known as frames. But MFCC is not robust to noise. Consequently, with different train and test conditions the accuracy of speech recognition systems decreases. On the other hand, using MFCCs of larger window of frames in GMMs needs more computational power. In this paper, Deep Belief Networks... 

In this paper, we address the problem of recovering degraded images using multivariate Gaussian mixture model (GMM) as a prior. The GMM framework in our method for image restoration is based on the assumption that the accumulation of similar patches in a neighborhood are derived from a multivariate Gaussian probability distribution with a specific covariance and mean. Previous methods of image restoration with GMM have not considered spatial (geometric) distance between patches in clustering. Our conducted experiments show that in the case of constraining Gaussian estimates into a finite-sized windows, the patch clusters are more likely to be derived from the estimated multivariate Gaussian... 

In this paper we address the problem of image interpolation using Gaussian Mixture Models (GMM) as a prior. Previous methods of image restoration with GMM have not considered spatial (geometric) distance between patches in clustering, failing to fully exploit the coherency of nearby patches. The GMM framework in our method for image interpolation is based on the assumption that the accumulation of similar patches in a neighborhood are derived from a multivariate Gaussian probability distribution with a specific covariance and mean. An Expectation Maximization-like (EM-like) algorithm is used in order to determine patches in a cluster and restore them. The results show that our image... 

We have recently proposed a correlated model to provide a Gaussian mixture representation of the cardiovascular signals, with promising results in identifying rhythm disturbances. The approach provides a transformation of the data into a set of integrable Gaussians distributed over time. Looking into the model from a new joint modeling perspective, it is capable of assembling a filtered estimation, and can be used to derive temporal information of the waveforms. In this paper, we present a step-by-step derivation of the joint model putting correlation assumptions together to conclude a minimal joint description for a pair of ECG-ABP signals. We then probe novel applications of this model,... 

This study focuses on the mode classification of phones speaker modes using GMM 1. In this regard, speech data in both enabled and disabled speaker modes of cell phones and telephones were collected, processed and classified into two different categories. The different mixture numbers (1 to 4) of GMM and wave files sizes of 10, 20, 40 and 80 kb were tested in order to obtain an optimal condition for classification. The GMM method attained 87.99% correct classification rate on test data. This classification is important for speech enabled IVR 2 systems [1], dialog systems and many systems in speech processing in the sense that it could help to load an optimum model for increasing system... 

A probabilistic formulation for semantic image annotation and retrieval is proposed. Annotation and retrieval are posed as classification problems where each class is defined as the group of database images labeled with a common semantic label. It is shown that, by establishing this one-to-one correspondence between semantic labels and semantic classes, a minimum probability of error annotation and retrieval are feasible with algorithms that are 1) conceptually simple and 2) computationally efficient. In this article, a content-based image retrieval and annotation architecture is proposed. Its attitude is decreasing the semantic gap by partitioning the image to its semantic regions and using... 

This paper addresses a model-based audio content analysis for classification of speech-music mixed audio signals into speech and music. A set of new features is presented and evaluated based on sinusoidal modeling of audio signals. The new feature set, including variance of the birth frequencies and duration of the longest frequency track in sinusoidal model, as a measure of the harmony and signal continuity, is introduced and discussed in detail. These features are used and compared to typical features as inputs to an audio classifier. Performance of these sinusoidal model features is evaluated through classification of audio into speech and music using both the GMM (Gaussian Mixture Model)... 

In this paper, a set of features is presented and evaluated based on sinusoidal modeling of audio signals. Amplitude, frequency, and phase parameters of the sinusoidal model are used and compared as input features into an audio classifier system. The performance of sinusoidal model features is evaluated for classification of audio into speech and music classes using both the Gaussian and the GMM (Gaussian Mixture Model) classifiers. Experimental results show superiority of the amplitude parameters of the sinusoidal model, which could be used for the first time for such an audio classification, as compared to the popular cepstral features. By using a set of 40 sinusoidal features, we achieved... 

A new approach for person identification based on hand geometry is presented. After preprocessing hand features are extracted from a photograph taken while user has placed his/her hand (either left or right) on the platform of a document scanner with no limits or fixation. Different pattern recognition techniques like Gaussian mixture modeling (GMM), Radial basis function neural networks (RBF), Multi layer perceptron (MLP), k-Nearest Neighbor (k-NN), Bayes method and mahalanobis/Hamming distance have been used in classification section. Experimental results show a rate of success above 90%. © 2005 IEEE