Log-based sparse nonnegative matrix factorization for data representation

C Peng, Y Zhang, Y Chen, Z Kang, C Chen… - Knowledge-Based …, 2022 - Elsevier
Nonnegative matrix factorization (NMF) has been widely studied in recent years due to its
effectiveness in representing nonnegative data with parts-based representations. For NMF …

Rectified Gaussian scale mixtures and the sparse non-negative least squares problem

A Nalci, I Fedorov, M Al-Shoukairi… - IEEE Transactions on …, 2018 - ieeexplore.ieee.org
In this paper, we develop a Bayesian evidence maximization framework to solve the sparse
non-negative least squares (S-NNLS) problem. We introduce a family of probability densities …

Dictionaries in machine learning

K Kreutz-Delgado, B Rao, I Fedorov, S Das - Signal Processing and …, 2024 - Elsevier
Abstract Machine learning and artificial intelligence are domains that have long been
influenced by and, in turn, have influenced our understanding of how effective and efficient …

Image denoising neural network architecture and method of training the same

M El-Khamy, I Fedorov, J Lee - US Patent 10,726,525, 2020 - Google Patents
US10726525B2 - Image denoising neural network architecture and method of training the same
- Google Patents US10726525B2 - Image denoising neural network architecture and method …

Multimodal sparse bayesian dictionary learning

I Fedorov, BD Rao - arxiv preprint arxiv:1804.03740, 2018 - arxiv.org
This paper addresses the problem of learning dictionaries for multimodal datasets, ie
datasets collected from multiple data sources. We present an algorithm called multimodal …

Nonnegative sparse representation of signals with a determinant-type sparsity measure based on the dc programming

B Tan, S Ding, Y Li, X Li - IEEE Access, 2018 - ieeexplore.ieee.org
Nonnegative sparse representation has become highly popular in certain applications in the
context of signals and corresponding dictionaries that have nonnegative limitations …

[HTML][HTML] A novel framework for the NMF methods with experiments to unmixing signals and feature representation

Y Teng, Y Yao, S Qi, C Li, L Xu, W Qian, F Fan… - … of Computational and …, 2019 - Elsevier
Non-negative matrix factorization (NMF) can be used in clustering, feature representation or
blind source separation. Many NMF methods have been developed including least squares …

[LIVRE][B] Structured Learning with Scale Mixture Priors

I Fedorov - 2018 - search.proquest.com
Sparsity plays an essential role in a number of modern algorithms. This thesis examines
how we can incorporate additional structural information within the sparsity profile and …

[PDF][PDF] Non-Negative Matrix Factorization Meets Time-Inhomogeneous Markov Chains

I Redko, M Sebban, A Habrard - 2020 - opt-ml.org
Non-negative matrix factorization (NMF)[22] is a popular unsupervised learning approach
that allows to obtain part-based representations of non-negative data samples and provide …

Robustness verification and novelty detection of deep learning-based raman spectral Analysis

Z Lei - 2022 - era.ed.ac.uk
Sparse coding aims at forming a linear combination of a few basis atoms out of a large set,
to reconstruct input vectors. Conventional model-based methods often tend to be slow and …