Yoleidy Huérfano-Maldonado

Yoleidy Huérfano-Maldonado

Auto-generated profile, based on publicly available information

Yoleidy Huérfano-Maldonado

Universidad de los Andes

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 July 2023 in Neurocomputing
Verified
Journal Article
Closed access
YHYoleidy Huérfano-Maldonado
Yoleidy Huérfano-Maldonado
MMMarco Mora
Marco Mora
KVKarina Vilches
Karina Vilches
Ruber Hernández-GarcíaRuber Hernández-García
Ruber Hernández-García
RGRodrigo Gutiérrez
Rodrigo Gutiérrez
MVMiguel Vera
Miguel Vera

14 May 2025 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
YHYoleidy Huérfano-Maldonado
Yoleidy Huérfano-Maldonado
KVKarina Vilches
Karina Vilches
MMMarco Mora
Marco Mora
CTClovis Tauber
Clovis Tauber
MVMiguel Angel Vera
Miguel Angel Vera
Computational algorithms that utilize nondifferentiable functions have proven highly effective in machine learning applications. This study introduces a novel framework for incorporating nondifferentiable functions into the objective functions of random-weight neural networks, specifically focusing on functional link random vector functional-link (RVFL) networks and extreme learning machines (ELMs). Our framework explores six nondifferentiable functions: the norms $ L_{1,1}$ , $ L_{1,2}$ , and $ L_{2,2}$ and the functions AbsMin, AbsMax, and a seminorm MaxMin. To enhance robustness, Fourier random assignments are applied as activation functions within these networks. The integration of these nondifferentiable functions into the objective functions of RVFL and ELM aims to reduce computational time in both training and testing stages, without compromising accuracy. We conducted extensive experiments on 12 benchmark datasets, encompassing small, medium, and large datasets, to evaluate the proposed algorithms against the $L_{2,1}$ -regularized random Fourier feature ELM ( $L_{2,1}$ -RF-ELM), which uses joint-norm regularization ( $L_{r,p}$ ) as documented in previous studies. Our findings indicate that the algorithms based on nondifferentiable functions not only achieve high accuracy but also significantly reduce computation time compared to the $L_{2,1}$ -based algorithm and other standard machine learning approaches.

1 May 2020 in Revista de Salud Pública
Verified
Journal Article
Open access
PDF file
ABAntonio J Bravo
Antonio J Bravo
MVMiguel Angel Vera
Miguel Angel Vera
YHYoleify K Huérfano
Yoleify K Huérfano
Objetivo Estimar el comportamiento de la infección por COVID-19 en Colombia mediante modelos matemáticos.Métodos Se construyeron dos modelos matemáticos para estimar los casos confirmados importados y los casos confirmados relacionados de la infección por COVID-19 en Colombia, respectivamente. La fenomenología de los casos confirmados importados es modelada con una función sigmoidal, mientras que los casos confirmados relacionados son modelados mediante una combinación de funciones exponenciales y funciones algebraicas polinomiales. Se utilizan algoritmos de ajuste basados en métodos de mínimos cuadrados y métodos de búsqueda directa para la determinación de los parámetros de los modelos.Resultados El modelo sigmodial realiza una estimación altamente convergente de los datos reportados, al 28 de mayo de 2020, de los casos confirmados importados de infección por COVID-19. El modelo muestra un error de predicción de 0,5%, que se mide usando la raíz del error cuadrático medio normalizado. El modelo para los casos confirmados reportados como relacionados muestra un error en la predicción del 3,5 % y un sesgo bajo del -0,01 asociado a la sobrestimación.Conclusiones El presente trabajo evidencia que los modelos matemáticos permiten eficaz y efectivamente predecir el comportamiento de la infección por COVID-19 en Colombia cuando los casos importados y los casos relacionados de infección son consideradores de manera independiente.

27 July 2020 in Revista Facultad de Ingeniería Universidad de Antioquia
Verified
Journal Article
Open access
PDF file
ABAntonio J Bravo
Antonio J Bravo
MVMiguel Angel Vera
Miguel Angel Vera
YMYoleidy Katherine Huérfano Maldonado
Yoleidy Katherine Huérfano Maldonado
YHYeison Fabián Manrique Hidalgo
Yeison Fabián Manrique Hidalgo
This research focuses on the study of a particular filter based on a similarity criterion that has been applied to improve the information contained in images acquired using different cardiac imaging modalities. The primary attention of this study is to examine which component of the similarity criterion generates more relevant information useful to increase the medical image quality. In this sense, four case studies are established, first a complete formulation of the similarity criterion is considered, and then three additional cases, representing each component of the criterion; such cases are referred to as full, main, residual1, and residual2, respectively. A score function is used for quantifying and then assessing the impact of each component of the similarity criterion. Such measure is a relation between some full–reference and blind–reference image enhancement measures. A computer generated phantom and a representative clinical dataset (1270 three–dimensional images from 126 patients) are used in a thorough evaluation of the similarity criterion. In general terms of performance of the image enhancement technique, the results of the study reveal that the component residual1 outperforms than the other two components of similarity criterion or its complete formulation.