Generalized Alpha-Beta Divergences and Their Application to Robust Nonnegative Matrix Factorization

Cichocki, Andrzej; Cruces, Sergio; Амари, Шун-ичи

doi:10.3390/e13010134

Cited by 175 publications

(181 citation statements)

References 52 publications

(50 reference statements)

Supporting

Mentioning

177

Contrasting

Unclassified

Order By: Relevance

“…Recently, the AB divergence has been proposed and its applications as a cost function for non-negative matrix factorization have been investigated (Cichocki et al, 2011). Motivated by its capabilities to weight and scale the individual ratios of the noisy speech and its approximation, y i l /ŷ i c,l whereŷ c,l = A c,l x c,l , we investigate the recognition performance of the proposed system using the AB divergence for d. The influence of different (α,β) values on this ratio is detailed in (Cichocki et al, 2011).…”

Section: Finding Exemplar Weightsmentioning

confidence: 99%

“…Motivated by its capabilities to weight and scale the individual ratios of the noisy speech and its approximation, y i l /ŷ i c,l whereŷ c,l = A c,l x c,l , we investigate the recognition performance of the proposed system using the AB divergence for d. The influence of different (α,β) values on this ratio is detailed in (Cichocki et al, 2011).…”

Section: Finding Exemplar Weightsmentioning

confidence: 99%

“…The two parameters of the AB divergence can be automatically adjusted based on the amount of contamination in the target utterance as the recognition performance for different noise levels depends on the emphasized (reliable) time-frequency bins. For the NSC solution, we apply the relaxation and non-linear projection techniques proposed in (Cichocki et al, 2006) for faster convergence to the multiplicative update rule derived in (Cichocki et al, 2011) to minimize the approximation error. The multiplicative update rule which minimizes the approximation error using the AB divergence for α 0 is given by…”

Section: Finding Exemplar Weightsmentioning

confidence: 99%

“…The optimal divergence is expected to weight the individual reconstruction errors of each time-frequency cells in a way that the most informative cells contribute the most to the total reconstruction error. In this work, we use the Alpha-Beta (AB) divergence (Cichocki et al, 2011) in place of the generalized KLD to quantify the approximation error. The AB divergence is a family of divergences with two parameters, namely α and β.…”

Section: Introductionmentioning

confidence: 99%

“…For different values of these parameters, the AB divergence connects various well-known distance/divergence measures such as the squared Euclidean distance, Hellinger distance, Itakura-Saito divergence and generalized KLD. The higher degree of freedom offered by the AB divergence has been shown to enable better robustness against noise and outliers (Cichocki et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Noise robust exemplar matching with alpha–beta divergence

Yılmaz

Gemmeke

hamme

2016

Speech Communication

View full text Add to dashboard Cite

The noise robust exemplar matching (N-REM) framework performs automatic speech recognition using exemplars, which are the labeled spectrographic representations of speech segments extracted from training data. By incorporating a sparse representations formulation, this technique remedies the inherent noise modeling problem of conventional exemplar matching-based automatic speech recognition systems. In this framework, noisy speech segments are approximated as a sparse linear combination of the exemplars of multiple lengths, each associated with a single speech unit such as words, half-words or phones. On account of the reconstruction error-based back end, the recognition accuracy highly depends on the congruence of the speech features and the divergence metric used to compare the speech segments with exemplars. In this work, we replace the conventional KullbackLeibler divergence (KLD) with a generalized divergence family called the Alpha-Beta divergence with two parameters, α and β, in conjunction with mel-scaled magnitude spectral features. The proposed recognizer traverses the (α,β) plane depending on the amount of contamination to provide better separation of speech and noise sources. Moreover, we apply our recently proposed active noise exemplar selection (ANES) technique in a more realistic scenario where the target utterances are degraded by genuine room noise. Recognition experiments on the small vocabulary track of the 2 nd CHiME Challenge and the AURORA-2 database have shown that the novel recognizer with the AB divergence and ANES outperforms the baseline system using the generalized KLD with tuned sparsity, especially at lower SNR levels.

show abstract

Section: Finding Exemplar Weightsmentioning

confidence: 99%

Section: Finding Exemplar Weightsmentioning

confidence: 99%

Section: Finding Exemplar Weightsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Noise robust exemplar matching with alpha–beta divergence

Yılmaz

Gemmeke

hamme

2016

Speech Communication

View full text Add to dashboard Cite

show abstract

A comprehensive review of generative adversarial networks: Fundamentals, applications, and challenges

Mohammed

2023

WIREs Computational Stats

View full text Add to dashboard Cite

In machine learning, a generative model is responsible for generating new samples of data in terms of a probabilistic model. Generative adversarial network (GAN) has been widely used to generate realistic samples in different domains and outperforms its peers in the generative models family. However, producing a robust GAN model is not a trivial task because many challenges face the GAN during the training process and impact its performance, affecting the quality and diversity of the generated samples. In this article, we conduct a comprehensive review of GANs to present the fundamentals of GAN, including its components, types, and objective functions. Also, we present an overview of the evaluation matrices used to evaluate GAN models. Moreover, we list the applications of GANs and research work in various domains. Finally, we present the challenges that face GANs and highlight two significant issues, representing mode collapse and training instability, in addition to those research efforts that tackle these challenges.This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Deep Learning Statistical Learning and Exploratory Methods of the Data Sciences > Neural Networks

show abstract